mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-05 16:48:03 +08:00
[Doc] Add Python comments for external models (#408)
* first commit for yolov7 * pybind for yolov7 * CPP README.md * CPP README.md * modified yolov7.cc * README.md * python file modify * delete license in fastdeploy/ * repush the conflict part * README.md modified * README.md modified * file path modified * file path modified * file path modified * file path modified * file path modified * README modified * README modified * move some helpers to private * add examples for yolov7 * api.md modified * api.md modified * api.md modified * YOLOv7 * yolov7 release link * yolov7 release link * yolov7 release link * copyright * change some helpers to private * change variables to const and fix documents. * gitignore * Transfer some funtions to private member of class * Transfer some funtions to private member of class * Merge from develop (#9) * Fix compile problem in different python version (#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> * Add PaddleDetetion/PPYOLOE model support (#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> * first commit for yolor * for merge * Develop (#11) * Fix compile problem in different python version (#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> * Add PaddleDetetion/PPYOLOE model support (#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> * Yolor (#16) * Develop (#11) (#12) * Fix compile problem in different python version (#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> * Add PaddleDetetion/PPYOLOE model support (#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> * Develop (#13) * Fix compile problem in different python version (#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> * Add PaddleDetetion/PPYOLOE model support (#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> * documents * documents * documents * documents * documents * documents * documents * documents * documents * documents * documents * documents * Develop (#14) * Fix compile problem in different python version (#26) * fix some usage problem in linux * Fix compile problem Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> * Add PaddleDetetion/PPYOLOE model support (#22) * add ppdet/ppyoloe * Add demo code and documents * add convert processor to vision (#27) * update .gitignore * Added checking for cmake include dir * fixed missing trt_backend option bug when init from trt * remove un-need data layout and add pre-check for dtype * changed RGB2BRG to BGR2RGB in ppcls model * add model_zoo yolov6 c++/python demo * fixed CMakeLists.txt typos * update yolov6 cpp/README.md * add yolox c++/pybind and model_zoo demo * move some helpers to private * fixed CMakeLists.txt typos * add normalize with alpha and beta * add version notes for yolov5/yolov6/yolox * add copyright to yolov5.cc * revert normalize * fixed some bugs in yolox * fixed examples/CMakeLists.txt to avoid conflicts * add convert processor to vision * format examples/CMakeLists summary * Fix bug while the inference result is empty with YOLOv5 (#29) * Add multi-label function for yolov5 * Update README.md Update doc * Update fastdeploy_runtime.cc fix variable option.trt_max_shape wrong name * Update runtime_option.md Update resnet model dynamic shape setting name from images to x * Fix bug when inference result boxes are empty * Delete detection.py Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason <928090362@qq.com> * add is_dynamic for YOLO series (#22) * modify ppmatting backend and docs * modify ppmatting docs * fix the PPMatting size problem * fix LimitShort's log * retrigger ci * modify PPMatting docs * modify the way for dealing with LimitShort * add python comments for external models * modify resnet c++ comments * modify C++ comments for external models * modify python comments and add result class comments * fix comments compile error * modify result.h comments * modify yolor comments Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: root <root@bjyz-sys-gpu-kongming3.bjyz.baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason <928090362@qq.com>
This commit is contained in:
ziqi-jin
34
docs/api_docs/python/face_detection.md
Normal file
34
docs/api_docs/python/face_detection.md
Normal file
@@ -0,0 +1,34 @@
|
|||||||
|
# Face Detection API
|
||||||
|
|
||||||
|
## fastdeploy.vision.facedet.RetinaFace
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.facedet.RetinaFace
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
|
||||||
|
## fastdeploy.vision.facedet.SCRFD
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.facedet.SCRFD
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.facedet.UltraFace
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.facedet.UltraFace
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.facedet.YOLOv5Face
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.facedet.YOLOv5Face
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
41
docs/api_docs/python/face_recognition.md
Normal file
41
docs/api_docs/python/face_recognition.md
Normal file
@@ -0,0 +1,41 @@
|
|||||||
|
# Face Recognition API
|
||||||
|
|
||||||
|
## fastdeploy.vision.faceid.AdaFace
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.faceid.AdaFace
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.faceid.CosFace
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.faceid.CosFace
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.faceid.ArcFace
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.faceid.ArcFace
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.faceid.PartialFC
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.faceid.PartialFC
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.faceid.VPL
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.faceid.VPL
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
@@ -18,4 +18,6 @@ FastDeploy
|
|||||||
image_classification.md
|
image_classification.md
|
||||||
keypoint_detection.md
|
keypoint_detection.md
|
||||||
matting.md
|
matting.md
|
||||||
|
face_recognition.md
|
||||||
|
face_detection.md
|
||||||
vision_results_en.md
|
vision_results_en.md
|
||||||
|
|||||||
@@ -1,3 +1,17 @@
|
|||||||
# Matting API
|
# Matting API
|
||||||
|
|
||||||
comming soon...
|
## fastdeploy.vision.matting.MODNet
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.matting.MODNet
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.matting.PPMatting
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.matting.PPMatting
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|||||||
@@ -63,3 +63,93 @@
|
|||||||
:members:
|
:members:
|
||||||
:inherited-members:
|
:inherited-members:
|
||||||
```
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.NanoDetPlus
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.NanoDetPlus
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.ScaledYOLOv4
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.ScaledYOLOv4
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOR
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOR
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOv5
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOv5
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOv5Lite
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOv5Lite
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOv6
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOv6
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOv7
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOv7
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOR
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOR
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOv7End2EndORT
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOv7End2EndORT
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOv7End2EndTRT
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOv7End2EndTRT
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|
||||||
|
## fastdeploy.vision.detection.YOLOX
|
||||||
|
|
||||||
|
```{eval-rst}
|
||||||
|
.. autoclass:: fastdeploy.vision.detection.YOLOX
|
||||||
|
:members:
|
||||||
|
:inherited-members:
|
||||||
|
```
|
||||||
|
|||||||
@@ -25,7 +25,7 @@ namespace vision {
|
|||||||
*
|
*
|
||||||
*/
|
*/
|
||||||
namespace classification {
|
namespace classification {
|
||||||
/*! @brief ResNet series model
|
/*! @brief Torchvision ResNet series model
|
||||||
*/
|
*/
|
||||||
class FASTDEPLOY_DECL ResNet : public FastDeployModel {
|
class FASTDEPLOY_DECL ResNet : public FastDeployModel {
|
||||||
public:
|
public:
|
||||||
@@ -44,17 +44,18 @@ class FASTDEPLOY_DECL ResNet : public FastDeployModel {
|
|||||||
virtual std::string ModelName() const { return "ResNet"; }
|
virtual std::string ModelName() const { return "ResNet"; }
|
||||||
/** \brief Predict for the input "im", the result will be saved in "result".
|
/** \brief Predict for the input "im", the result will be saved in "result".
|
||||||
*
|
*
|
||||||
* \param[in] im Input image for inference.
|
* \param[in] im The input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format
|
||||||
* \param[in] result Saving the inference result.
|
* \param[in] result Saving the inference result.
|
||||||
* \param[in] topk The length of return values, e.g., if topk==2, the result will include the 2 most possible class label for input image.
|
* \param[in] topk The length of return values, e.g., if topk==2, the result will include the 2 most possible class label for input image.
|
||||||
*/
|
*/
|
||||||
virtual bool Predict(cv::Mat* im, ClassifyResult* result, int topk = 1);
|
virtual bool Predict(cv::Mat* im, ClassifyResult* result, int topk = 1);
|
||||||
|
/*! @brief
|
||||||
/// Tuple of (width, height)
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// Mean parameters for normalize
|
/// Mean parameters for normalize, size should be the the same as channels
|
||||||
std::vector<float> mean_vals;
|
std::vector<float> mean_vals;
|
||||||
/// Std parameters for normalize
|
/// Std parameters for normalize, size should be the the same as channels
|
||||||
std::vector<float> std_vals;
|
std::vector<float> std_vals;
|
||||||
|
|
||||||
|
|
||||||
|
|||||||
@@ -154,88 +154,117 @@ struct FASTDEPLOY_DECL OCRResult : public BaseResult {
|
|||||||
std::string Str();
|
std::string Str();
|
||||||
};
|
};
|
||||||
|
|
||||||
|
/*! @brief Face detection result structure for all the face detection models
|
||||||
|
*/
|
||||||
struct FASTDEPLOY_DECL FaceDetectionResult : public BaseResult {
|
struct FASTDEPLOY_DECL FaceDetectionResult : public BaseResult {
|
||||||
// box: xmin, ymin, xmax, ymax
|
/** \brief All the detected object boxes for an input image, the size of `boxes` is the number of detected objects, and the element of `boxes` is a array of 4 float values, means [xmin, ymin, xmax, ymax]
|
||||||
|
*/
|
||||||
std::vector<std::array<float, 4>> boxes;
|
std::vector<std::array<float, 4>> boxes;
|
||||||
// landmark: x, y, landmarks may empty if the
|
/** \brief
|
||||||
// model don't detect face with landmarks.
|
* If the model detect face with landmarks, every detected object box correspoing to a landmark, which is a array of 2 float values, means location [x,y]
|
||||||
// Note, one face might have multiple landmarks,
|
*/
|
||||||
// such as 5/19/21/68/98/..., etc.
|
|
||||||
std::vector<std::array<float, 2>> landmarks;
|
std::vector<std::array<float, 2>> landmarks;
|
||||||
|
/** \brief
|
||||||
|
* Indicates the confidence of all targets detected from a single image, and the number of elements is consistent with boxes.size()
|
||||||
|
*/
|
||||||
std::vector<float> scores;
|
std::vector<float> scores;
|
||||||
ResultType type = ResultType::FACE_DETECTION;
|
ResultType type = ResultType::FACE_DETECTION;
|
||||||
// set landmarks_per_face manually in your post processes.
|
/** \brief
|
||||||
|
* `landmarks_per_face` indicates the number of face landmarks for each detected face
|
||||||
|
* if the model's output contains face landmarks (such as YOLOv5Face, SCRFD, ...)
|
||||||
|
*/
|
||||||
int landmarks_per_face;
|
int landmarks_per_face;
|
||||||
|
|
||||||
FaceDetectionResult() { landmarks_per_face = 0; }
|
FaceDetectionResult() { landmarks_per_face = 0; }
|
||||||
FaceDetectionResult(const FaceDetectionResult& res);
|
FaceDetectionResult(const FaceDetectionResult& res);
|
||||||
|
/// Clear detection result
|
||||||
void Clear();
|
void Clear();
|
||||||
|
|
||||||
void Reserve(int size);
|
void Reserve(int size);
|
||||||
|
|
||||||
void Resize(int size);
|
void Resize(int size);
|
||||||
|
/// Debug function, convert the result to string to print
|
||||||
std::string Str();
|
std::string Str();
|
||||||
};
|
};
|
||||||
|
|
||||||
|
/*! @brief Segmentation result structure for all the segmentation models
|
||||||
|
*/
|
||||||
struct FASTDEPLOY_DECL SegmentationResult : public BaseResult {
|
struct FASTDEPLOY_DECL SegmentationResult : public BaseResult {
|
||||||
// mask
|
/** \brief
|
||||||
|
* `label_map` stores the pixel-level category labels for input image. the number of pixels is equal to label_map.size()
|
||||||
|
*/
|
||||||
std::vector<uint8_t> label_map;
|
std::vector<uint8_t> label_map;
|
||||||
|
/** \brief
|
||||||
|
* `score_map` stores the probability of the predicted label for each pixel of input image.
|
||||||
|
*/
|
||||||
std::vector<float> score_map;
|
std::vector<float> score_map;
|
||||||
|
/// The output shape, means [H, W]
|
||||||
std::vector<int64_t> shape;
|
std::vector<int64_t> shape;
|
||||||
bool contain_score_map = false;
|
bool contain_score_map = false;
|
||||||
|
|
||||||
ResultType type = ResultType::SEGMENTATION;
|
ResultType type = ResultType::SEGMENTATION;
|
||||||
|
/// Clear detection result
|
||||||
void Clear();
|
void Clear();
|
||||||
|
|
||||||
void Reserve(int size);
|
void Reserve(int size);
|
||||||
|
|
||||||
void Resize(int size);
|
void Resize(int size);
|
||||||
|
/// Debug function, convert the result to string to print
|
||||||
std::string Str();
|
std::string Str();
|
||||||
};
|
};
|
||||||
|
|
||||||
|
/*! @brief Face recognition result structure for all the Face recognition models
|
||||||
|
*/
|
||||||
struct FASTDEPLOY_DECL FaceRecognitionResult : public BaseResult {
|
struct FASTDEPLOY_DECL FaceRecognitionResult : public BaseResult {
|
||||||
// face embedding vector with 128/256/512 ... dim
|
/** \brief The feature embedding that represents the final extraction of the face recognition model can be used to calculate the feature similarity between faces.
|
||||||
|
*/
|
||||||
std::vector<float> embedding;
|
std::vector<float> embedding;
|
||||||
|
|
||||||
ResultType type = ResultType::FACE_RECOGNITION;
|
ResultType type = ResultType::FACE_RECOGNITION;
|
||||||
|
|
||||||
FaceRecognitionResult() {}
|
FaceRecognitionResult() {}
|
||||||
FaceRecognitionResult(const FaceRecognitionResult& res);
|
FaceRecognitionResult(const FaceRecognitionResult& res);
|
||||||
|
/// Clear detection result
|
||||||
void Clear();
|
void Clear();
|
||||||
|
|
||||||
void Reserve(int size);
|
void Reserve(int size);
|
||||||
|
|
||||||
void Resize(int size);
|
void Resize(int size);
|
||||||
|
/// Debug function, convert the result to string to print
|
||||||
std::string Str();
|
std::string Str();
|
||||||
};
|
};
|
||||||
|
|
||||||
|
/*! @brief Matting result structure for all the Matting models
|
||||||
|
*/
|
||||||
struct FASTDEPLOY_DECL MattingResult : public BaseResult {
|
struct FASTDEPLOY_DECL MattingResult : public BaseResult {
|
||||||
// alpha matte and fgr (predicted foreground: HWC/BGR float32)
|
/** \brief
|
||||||
|
`alpha` is a one-dimensional vector, which is the predicted alpha transparency value. The range of values is [0., 1.], and the length is hxw. h, w are the height and width of the input image
|
||||||
|
*/
|
||||||
std::vector<float> alpha; // h x w
|
std::vector<float> alpha; // h x w
|
||||||
|
/** \brief
|
||||||
|
If the model can predict foreground, `foreground` save the predicted foreground image, the shape is [hight,width,channel] generally.
|
||||||
|
*/
|
||||||
std::vector<float> foreground; // h x w x c (c=3 default)
|
std::vector<float> foreground; // h x w x c (c=3 default)
|
||||||
// height, width, channel for foreground and alpha
|
/** \brief
|
||||||
// must be (h,w,c) and setup before Reserve and Resize
|
* The shape of output result, when contain_foreground == false, shape only contains (h, w), when contain_foreground == true, shape contains (h, w, c), and c is generally 3
|
||||||
// c is only for foreground if contain_foreground is true.
|
*/
|
||||||
std::vector<int64_t> shape;
|
std::vector<int64_t> shape;
|
||||||
|
/** \brief
|
||||||
|
If the model can predict alpha matte and foreground, contain_foreground = true, default false
|
||||||
|
*/
|
||||||
bool contain_foreground = false;
|
bool contain_foreground = false;
|
||||||
|
|
||||||
ResultType type = ResultType::MATTING;
|
ResultType type = ResultType::MATTING;
|
||||||
|
|
||||||
MattingResult() {}
|
MattingResult() {}
|
||||||
MattingResult(const MattingResult& res);
|
MattingResult(const MattingResult& res);
|
||||||
|
/// Clear detection result
|
||||||
void Clear();
|
void Clear();
|
||||||
|
|
||||||
void Reserve(int size);
|
void Reserve(int size);
|
||||||
|
|
||||||
void Resize(int size);
|
void Resize(int size);
|
||||||
|
/// Debug function, convert the result to string to print
|
||||||
std::string Str();
|
std::string Str();
|
||||||
};
|
};
|
||||||
|
|
||||||
|
|||||||
@@ -53,21 +53,23 @@ class FASTDEPLOY_DECL NanoDetPlus : public FastDeployModel {
|
|||||||
float conf_threshold = 0.35f,
|
float conf_threshold = 0.35f,
|
||||||
float nms_iou_threshold = 0.5f);
|
float nms_iou_threshold = 0.5f);
|
||||||
|
|
||||||
/// tuple of input size (width, height), e.g (320, 320)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of input size (width, height), e.g (320, 320)
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/*! @brief
|
// keep aspect ratio or not when perform resize operation.
|
||||||
keep aspect ratio or not when perform resize operation. This option is set as `false` by default in NanoDet-Plus
|
// This option is set as `false` by default in NanoDet-Plus
|
||||||
*/
|
|
||||||
bool keep_ratio;
|
bool keep_ratio;
|
||||||
/*! @brief
|
// downsample strides for NanoDet-Plus to generate anchors,
|
||||||
downsample strides for NanoDet-Plus to generate anchors, will take (8, 16, 32, 64) as default values
|
// will take (8, 16, 32, 64) as default values
|
||||||
*/
|
|
||||||
std::vector<int> downsample_strides;
|
std::vector<int> downsample_strides;
|
||||||
/// for offseting the boxes by classes when using NMS, default 4096
|
// for offseting the boxes by classes when using NMS, default 4096
|
||||||
float max_wh;
|
float max_wh;
|
||||||
/// reg_max for GFL regression, default 7
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, reg_max for GFL regression, default 7
|
||||||
|
*/
|
||||||
int reg_max;
|
int reg_max;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -50,23 +50,23 @@ class FASTDEPLOY_DECL ScaledYOLOv4 : public FastDeployModel {
|
|||||||
float conf_threshold = 0.25,
|
float conf_threshold = 0.25,
|
||||||
float nms_iou_threshold = 0.5);
|
float nms_iou_threshold = 0.5);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/// for offseting the boxes by classes when using NMS
|
// for offseting the boxes by classes when using NMS
|
||||||
float max_wh;
|
float max_wh;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -1,4 +1,5 @@
|
|||||||
// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved. //NOLINT
|
|
||||||
|
// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
|
||||||
//
|
//
|
||||||
// Licensed under the Apache License, Version 2.0 (the "License");
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
||||||
// you may not use this file except in compliance with the License.
|
// you may not use this file except in compliance with the License.
|
||||||
@@ -48,23 +49,24 @@ class FASTDEPLOY_DECL YOLOR : public FastDeployModel {
|
|||||||
float conf_threshold = 0.25,
|
float conf_threshold = 0.25,
|
||||||
float nms_iou_threshold = 0.5);
|
float nms_iou_threshold = 0.5);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/// for offseting the boxes by classes when using NMS
|
// for offseting the boxes by classes when using NMS
|
||||||
float max_wh;
|
float max_wh;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -77,23 +77,24 @@ class FASTDEPLOY_DECL YOLOv5 : public FastDeployModel {
|
|||||||
float conf_threshold, float nms_iou_threshold, bool multi_label,
|
float conf_threshold, float nms_iou_threshold, bool multi_label,
|
||||||
float max_wh = 7680.0);
|
float max_wh = 7680.0);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size_;
|
std::vector<int> size_;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value_;
|
std::vector<float> padding_value_;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad_;
|
bool is_mini_pad_;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad_;
|
bool is_no_pad_;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up_;
|
bool is_scale_up_;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride_;
|
int stride_;
|
||||||
/// for offseting the boxes by classes when using NMS
|
// for offseting the boxes by classes when using NMS
|
||||||
float max_wh_;
|
float max_wh_;
|
||||||
/// for different strategies to get boxes when postprocessing
|
/// for different strategies to get boxes when postprocessing
|
||||||
bool multi_label_;
|
bool multi_label_;
|
||||||
|
|||||||
@@ -53,31 +53,30 @@ class FASTDEPLOY_DECL YOLOv5Lite : public FastDeployModel {
|
|||||||
|
|
||||||
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/// for offseting the boxes by classes when using NMS
|
// for offseting the boxes by classes when using NMS
|
||||||
float max_wh;
|
float max_wh;
|
||||||
/*! @brief
|
// downsample strides for YOLOv5Lite to generate anchors,
|
||||||
downsample strides for YOLOv5Lite to generate anchors, will take (8,16,32) as default values, might have stride=64.
|
// will take (8,16,32) as default values, might have stride=64.
|
||||||
*/
|
|
||||||
std::vector<int> downsample_strides;
|
std::vector<int> downsample_strides;
|
||||||
/*! @brief
|
// anchors parameters, downsample_strides will take (8,16,32),
|
||||||
anchors parameters, downsample_strides will take (8,16,32), each stride has three anchors with width and hight
|
// each stride has three anchors with width and hight
|
||||||
*/
|
|
||||||
std::vector<std::vector<float>> anchor_config;
|
std::vector<std::vector<float>> anchor_config;
|
||||||
/*! @brief
|
/*! @brief
|
||||||
whether the model_file was exported with decode module. The official
|
whether the model_file was exported with decode module. The official
|
||||||
|
|||||||
@@ -31,10 +31,12 @@ void BindYOLOv5Lite(pybind11::module& m) {
|
|||||||
.def("use_cuda_preprocessing",
|
.def("use_cuda_preprocessing",
|
||||||
[](vision::detection::YOLOv5Lite& self, int max_image_size) {
|
[](vision::detection::YOLOv5Lite& self, int max_image_size) {
|
||||||
self.UseCudaPreprocessing(max_image_size);
|
self.UseCudaPreprocessing(max_image_size);
|
||||||
})
|
})
|
||||||
.def_readwrite("size", &vision::detection::YOLOv5Lite::size)
|
.def_readwrite("size", &vision::detection::YOLOv5Lite::size)
|
||||||
.def_readwrite("padding_value",
|
.def_readwrite("padding_value",
|
||||||
&vision::detection::YOLOv5Lite::padding_value)
|
&vision::detection::YOLOv5Lite::padding_value)
|
||||||
|
.def_readwrite("downsample_strides",
|
||||||
|
&vision::detection::YOLOv5Lite::downsample_strides)
|
||||||
.def_readwrite("is_mini_pad", &vision::detection::YOLOv5Lite::is_mini_pad)
|
.def_readwrite("is_mini_pad", &vision::detection::YOLOv5Lite::is_mini_pad)
|
||||||
.def_readwrite("is_no_pad", &vision::detection::YOLOv5Lite::is_no_pad)
|
.def_readwrite("is_no_pad", &vision::detection::YOLOv5Lite::is_no_pad)
|
||||||
.def_readwrite("is_scale_up", &vision::detection::YOLOv5Lite::is_scale_up)
|
.def_readwrite("is_scale_up", &vision::detection::YOLOv5Lite::is_scale_up)
|
||||||
|
|||||||
@@ -56,25 +56,25 @@ class FASTDEPLOY_DECL YOLOv6 : public FastDeployModel {
|
|||||||
|
|
||||||
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/*! @brief
|
// for offseting the boxes by classes when using NMS,
|
||||||
for offseting the boxes by classes when using NMS, default 4096 in meituan/YOLOv6
|
// default 4096 in meituan/YOLOv6
|
||||||
*/
|
|
||||||
float max_wh;
|
float max_wh;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -53,23 +53,24 @@ class FASTDEPLOY_DECL YOLOv7 : public FastDeployModel {
|
|||||||
|
|
||||||
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/// for offseting the boxes by classes when using NMS
|
// for offseting the boxes by classes when using NMS
|
||||||
float max_wh;
|
float max_wh;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -47,21 +47,22 @@ class FASTDEPLOY_DECL YOLOv7End2EndORT : public FastDeployModel {
|
|||||||
virtual bool Predict(cv::Mat* im, DetectionResult* result,
|
virtual bool Predict(cv::Mat* im, DetectionResult* result,
|
||||||
float conf_threshold = 0.25);
|
float conf_threshold = 0.25);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -52,21 +52,22 @@ class FASTDEPLOY_DECL YOLOv7End2EndTRT : public FastDeployModel {
|
|||||||
|
|
||||||
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
void UseCudaPreprocessing(int max_img_size = 3840 * 2160);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -51,9 +51,11 @@ class FASTDEPLOY_DECL YOLOX : public FastDeployModel {
|
|||||||
float conf_threshold = 0.25,
|
float conf_threshold = 0.25,
|
||||||
float nms_iou_threshold = 0.5);
|
float nms_iou_threshold = 0.5);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/*! @brief
|
/*! @brief
|
||||||
whether the model_file was exported with decode module. The official
|
whether the model_file was exported with decode module. The official
|
||||||
@@ -62,11 +64,10 @@ class FASTDEPLOY_DECL YOLOX : public FastDeployModel {
|
|||||||
was exported with decode module.
|
was exported with decode module.
|
||||||
*/
|
*/
|
||||||
bool is_decode_exported;
|
bool is_decode_exported;
|
||||||
/*! @brief
|
// downsample strides for YOLOX to generate anchors,
|
||||||
downsample strides for YOLOX to generate anchors, will take (8,16,32) as default values, might have stride=64
|
// will take (8,16,32) as default values, might have stride=64
|
||||||
*/
|
|
||||||
std::vector<int> downsample_strides;
|
std::vector<int> downsample_strides;
|
||||||
/// for offseting the boxes by classes when using NMS, default 4096
|
// for offseting the boxes by classes when using NMS, default 4096
|
||||||
float max_wh;
|
float max_wh;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -52,19 +52,25 @@ class FASTDEPLOY_DECL RetinaFace : public FastDeployModel {
|
|||||||
float conf_threshold = 0.25f,
|
float conf_threshold = 0.25f,
|
||||||
float nms_iou_threshold = 0.4f);
|
float nms_iou_threshold = 0.4f);
|
||||||
|
|
||||||
/// tuple of (width, height), default (640, 640)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize, default (640, 640)
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/*! @brief
|
/*! @brief
|
||||||
variance in RetinaFace's prior-box(anchor) generate process, default (0.1, 0.2)
|
Argument for image postprocessing step, variance in RetinaFace's prior-box(anchor) generate process, default (0.1, 0.2)
|
||||||
*/
|
*/
|
||||||
std::vector<float> variance;
|
std::vector<float> variance;
|
||||||
/*! @brief
|
/*! @brief
|
||||||
downsample strides (namely, steps) for RetinaFace to generate anchors, will take (8,16,32) as default values
|
Argument for image postprocessing step, downsample strides (namely, steps) for RetinaFace to generate anchors, will take (8,16,32) as default values
|
||||||
*/
|
*/
|
||||||
std::vector<int> downsample_strides;
|
std::vector<int> downsample_strides;
|
||||||
/// min sizes, width and height for each anchor
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, min sizes, width and height for each anchor
|
||||||
|
*/
|
||||||
std::vector<std::vector<int>> min_sizes;
|
std::vector<std::vector<int>> min_sizes;
|
||||||
/// landmarks_per_face, default 5 in RetinaFace
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, landmarks_per_face, default 5 in RetinaFace
|
||||||
|
*/
|
||||||
int landmarks_per_face;
|
int landmarks_per_face;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -51,33 +51,40 @@ class FASTDEPLOY_DECL SCRFD : public FastDeployModel {
|
|||||||
float conf_threshold = 0.25f,
|
float conf_threshold = 0.25f,
|
||||||
float nms_iou_threshold = 0.4f);
|
float nms_iou_threshold = 0.4f);
|
||||||
|
|
||||||
/// tuple of (width, height), default (640, 640)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize, default (640, 640)
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// padding value, size should be the same as channels
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
std::vector<float> padding_value;
|
std::vector<float> padding_value;
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
bool is_mini_pad;
|
bool is_mini_pad;
|
||||||
/*! @brief
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
// will resize the image to the set size
|
||||||
*/
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/*! @brief
|
/*! @brief
|
||||||
downsample strides (namely, steps) for SCRFD to generate anchors, will take (8,16,32) as default values
|
Argument for image postprocessing step, downsample strides (namely, steps) for SCRFD to generate anchors, will take (8,16,32) as default values
|
||||||
*/
|
*/
|
||||||
std::vector<int> downsample_strides;
|
std::vector<int> downsample_strides;
|
||||||
/// landmarks_per_face, default 5 in SCRFD
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, landmarks_per_face, default 5 in SCRFD
|
||||||
|
*/
|
||||||
int landmarks_per_face;
|
int landmarks_per_face;
|
||||||
/// the outputs of onnx file with key points features or not
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, the outputs of onnx file with key points features or not
|
||||||
|
*/
|
||||||
bool use_kps;
|
bool use_kps;
|
||||||
/// the upperbond number of boxes processed by nms
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, the upperbond number of boxes processed by nms
|
||||||
|
*/
|
||||||
int max_nms;
|
int max_nms;
|
||||||
/// number anchors of each stride
|
/// Argument for image postprocessing step, anchor number of each stride
|
||||||
unsigned int num_anchors;
|
unsigned int num_anchors;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -52,7 +52,9 @@ class FASTDEPLOY_DECL UltraFace : public FastDeployModel {
|
|||||||
float conf_threshold = 0.7f,
|
float conf_threshold = 0.7f,
|
||||||
float nms_iou_threshold = 0.3f);
|
float nms_iou_threshold = 0.3f);
|
||||||
|
|
||||||
/// tuple of (width, height), default (320, 240)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize, default (320, 240)
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
|
|
||||||
private:
|
private:
|
||||||
|
|||||||
@@ -50,26 +50,27 @@ class FASTDEPLOY_DECL YOLOv5Face : public FastDeployModel {
|
|||||||
float conf_threshold = 0.25,
|
float conf_threshold = 0.25,
|
||||||
float nms_iou_threshold = 0.5);
|
float nms_iou_threshold = 0.5);
|
||||||
|
|
||||||
/// tuple of (width, height)
|
|
||||||
std::vector<int> size;
|
|
||||||
/// padding value, size should be the same as channels
|
|
||||||
std::vector<float> padding_value;
|
|
||||||
/// only pad to the minimum rectange which height and width is times of stride
|
|
||||||
bool is_mini_pad;
|
|
||||||
/*! @brief
|
/*! @brief
|
||||||
while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize
|
||||||
*/
|
*/
|
||||||
|
std::vector<int> size;
|
||||||
|
// padding value, size should be the same as channels
|
||||||
|
|
||||||
|
std::vector<float> padding_value;
|
||||||
|
// only pad to the minimum rectange which height and width is times of stride
|
||||||
|
bool is_mini_pad;
|
||||||
|
// while is_mini_pad = false and is_no_pad = true,
|
||||||
|
// will resize the image to the set size
|
||||||
|
|
||||||
bool is_no_pad;
|
bool is_no_pad;
|
||||||
/*! @brief
|
// if is_scale_up is false, the input image only can be zoom out,
|
||||||
if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
// the maximum resize scale cannot exceed 1.0
|
||||||
*/
|
|
||||||
|
|
||||||
bool is_scale_up;
|
bool is_scale_up;
|
||||||
/// padding stride, for is_mini_pad
|
// padding stride, for is_mini_pad
|
||||||
int stride;
|
int stride;
|
||||||
/*! @brief
|
/*! @brief
|
||||||
setup the number of landmarks for per face (if have), default 5 in
|
Argument for image postprocessing step, setup the number of landmarks for per face (if have), default 5 in
|
||||||
official yolov5face note that, the outupt tensor's shape must be:
|
official yolov5face note that, the outupt tensor's shape must be:
|
||||||
(1,n,4+1+2*landmarks_per_face+1=box+obj+landmarks+cls)
|
(1,n,4+1+2*landmarks_per_face+1=box+obj+landmarks+cls)
|
||||||
*/
|
*/
|
||||||
|
|||||||
@@ -40,15 +40,21 @@ class FASTDEPLOY_DECL InsightFaceRecognitionModel : public FastDeployModel {
|
|||||||
|
|
||||||
virtual std::string ModelName() const { return "deepinsight/insightface"; }
|
virtual std::string ModelName() const { return "deepinsight/insightface"; }
|
||||||
|
|
||||||
/// tuple of (width, height), default (112, 112)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize, default (112, 112)
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// alpha values for normalization
|
/// Argument for image preprocessing step, alpha values for normalization
|
||||||
std::vector<float> alpha;
|
std::vector<float> alpha;
|
||||||
/// beta values for normalization
|
/// Argument for image preprocessing step, beta values for normalization
|
||||||
std::vector<float> beta;
|
std::vector<float> beta;
|
||||||
/// whether to swap the B and R channel, such as BGR->RGB, default true.
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
*/
|
||||||
bool swap_rb;
|
bool swap_rb;
|
||||||
/// whether to apply l2 normalize to embedding values, default;
|
/*! @brief
|
||||||
|
Argument for image postprocessing step, whether to apply l2 normalize to embedding values, default false;
|
||||||
|
*/
|
||||||
bool l2_normalize;
|
bool l2_normalize;
|
||||||
/** \brief Predict the face recognition result for an input image
|
/** \brief Predict the face recognition result for an input image
|
||||||
*
|
*
|
||||||
|
|||||||
@@ -39,13 +39,21 @@ class FASTDEPLOY_DECL MODNet : public FastDeployModel {
|
|||||||
|
|
||||||
std::string ModelName() const { return "matting/MODNet"; }
|
std::string ModelName() const { return "matting/MODNet"; }
|
||||||
|
|
||||||
/// tuple of (width, height), default (256, 256)
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, tuple of (width, height), decide the target size after resize, default (256, 256)
|
||||||
|
*/
|
||||||
std::vector<int> size;
|
std::vector<int> size;
|
||||||
/// parameters for normalization
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, parameters for normalization, size should be the the same as channels
|
||||||
|
*/
|
||||||
std::vector<float> alpha;
|
std::vector<float> alpha;
|
||||||
/// parameters for normalization
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, parameters for normalization, size should be the the same as channels
|
||||||
|
*/
|
||||||
std::vector<float> beta;
|
std::vector<float> beta;
|
||||||
/// whether to swap the B and R channel, such as BGR->RGB, default true.
|
/*! @brief
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
*/
|
||||||
bool swap_rb;
|
bool swap_rb;
|
||||||
/** \brief Predict the matting result for an input image
|
/** \brief Predict the matting result for an input image
|
||||||
*
|
*
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class NanoDetPlus(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a NanoDetPlus model exported by NanoDet.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./nanodet.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(NanoDetPlus, self).__init__(runtime_option)
|
super(NanoDetPlus, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class NanoDetPlus(FastDeployModel):
|
|||||||
assert self.initialized, "NanoDetPlus initialize failed."
|
assert self.initialized, "NanoDetPlus initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,26 +55,36 @@ class NanoDetPlus(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [416, 416]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [416, 416]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def keep_ratio(self):
|
def keep_ratio(self):
|
||||||
|
# keep aspect ratio or not when perform resize operation. This option is set as false by default in NanoDet-Plus
|
||||||
return self._model.keep_ratio
|
return self._model.keep_ratio
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def downsample_strides(self):
|
def downsample_strides(self):
|
||||||
|
# downsample strides for NanoDet-Plus to generate anchors, will take (8, 16, 32, 64) as default values
|
||||||
return self._model.downsample_strides
|
return self._model.downsample_strides
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS, default 4096
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def reg_max(self):
|
def reg_max(self):
|
||||||
|
"""
|
||||||
|
reg_max for GFL regression, default 7
|
||||||
|
"""
|
||||||
return self._model.reg_max
|
return self._model.reg_max
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class ScaledYOLOv4(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a ScaledYOLOv4 model exported by ScaledYOLOv4.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./scaled_yolov4.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(ScaledYOLOv4, self).__init__(runtime_option)
|
super(ScaledYOLOv4, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class ScaledYOLOv4(FastDeployModel):
|
|||||||
assert self.initialized, "ScaledYOLOv4 initialize failed."
|
assert self.initialized, "ScaledYOLOv4 initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,30 +55,39 @@ class ScaledYOLOv4(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -92,19 +115,21 @@ class ScaledYOLOv4(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@max_wh.setter
|
@max_wh.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOR(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOR model exported by YOLOR
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolor.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOR, self).__init__(runtime_option)
|
super(YOLOR, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class YOLOR(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOR initialize failed."
|
assert self.initialized, "YOLOR initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,30 +55,39 @@ class YOLOR(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -92,19 +115,21 @@ class YOLOR(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@max_wh.setter
|
@max_wh.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv5(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv5 model exported by YOLOv5.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov5.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv5, self).__init__(runtime_option)
|
super(YOLOv5, self).__init__(runtime_option)
|
||||||
@@ -34,12 +41,16 @@ class YOLOv5(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv5 initialize failed."
|
assert self.initialized, "YOLOv5 initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
def use_cuda_preprocessing(self, max_image_size=3840 * 2160):
|
|
||||||
return self._model.use_cuda_preprocessing(max_image_size)
|
|
||||||
|
|
||||||
@staticmethod
|
@staticmethod
|
||||||
def preprocess(input_image,
|
def preprocess(input_image,
|
||||||
size=[640, 640],
|
size=[640, 640],
|
||||||
@@ -69,30 +80,39 @@ class YOLOv5(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@property
|
@property
|
||||||
@@ -124,19 +144,21 @@ class YOLOv5(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@max_wh.setter
|
@max_wh.setter
|
||||||
@@ -148,5 +170,6 @@ class YOLOv5(FastDeployModel):
|
|||||||
@multi_label.setter
|
@multi_label.setter
|
||||||
def multi_label(self, value):
|
def multi_label(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `multi_label` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `multi_label` must be type of bool."
|
||||||
self._model.multi_label = value
|
self._model.multi_label = value
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv5Lite(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv5Lite model exported by YOLOv5Lite.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov5lite.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv5Lite, self).__init__(runtime_option)
|
super(YOLOv5Lite, self).__init__(runtime_option)
|
||||||
@@ -34,50 +41,76 @@ class YOLOv5Lite(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv5Lite initialize failed."
|
assert self.initialized, "YOLOv5Lite initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
def use_cuda_preprocessing(self, max_image_size=3840 * 2160):
|
|
||||||
return self._model.use_cuda_preprocessing(max_image_size)
|
|
||||||
|
|
||||||
# 一些跟YOLOv5Lite模型有关的属性封装
|
# 一些跟YOLOv5Lite模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_decode_exported(self):
|
def is_decode_exported(self):
|
||||||
|
"""
|
||||||
|
whether the model_file was exported with decode module.
|
||||||
|
The official YOLOv5Lite/export.py script will export ONNX file without decode module.
|
||||||
|
Please set it 'true' manually if the model file was exported with decode module.
|
||||||
|
false : ONNX files without decode module. true : ONNX file with decode module.
|
||||||
|
"""
|
||||||
return self._model.is_decode_exported
|
return self._model.is_decode_exported
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def anchor_config(self):
|
def anchor_config(self):
|
||||||
return self._model.anchor_config
|
return self._model.anchor_config
|
||||||
|
|
||||||
|
@property
|
||||||
|
def downsample_strides(self):
|
||||||
|
"""
|
||||||
|
downsample strides for YOLOv5Lite to generate anchors, will take (8,16,32) as default values, might have stride=64.
|
||||||
|
"""
|
||||||
|
return self._model.downsample_strides
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
def size(self, wh):
|
def size(self, wh):
|
||||||
assert isinstance(wh, (list, tuple)),\
|
assert isinstance(wh, (list, tuple)),\
|
||||||
@@ -103,19 +136,21 @@ class YOLOv5Lite(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@max_wh.setter
|
@max_wh.setter
|
||||||
@@ -138,3 +173,10 @@ class YOLOv5Lite(FastDeployModel):
|
|||||||
assert isinstance(anchor_config_val[0], list),\
|
assert isinstance(anchor_config_val[0], list),\
|
||||||
"The value to set `anchor_config` must be 2-dimensions tuple or list"
|
"The value to set `anchor_config` must be 2-dimensions tuple or list"
|
||||||
self._model.anchor_config = anchor_config_val
|
self._model.anchor_config = anchor_config_val
|
||||||
|
|
||||||
|
@downsample_strides.setter
|
||||||
|
def downsample_strides(self, value):
|
||||||
|
assert isinstance(
|
||||||
|
value,
|
||||||
|
list), "The value to set `downsample_strides` must be type of list."
|
||||||
|
self._model.downsample_strides = value
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv6(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv6 model exported by YOLOv6.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov6.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv6, self).__init__(runtime_option)
|
super(YOLOv6, self).__init__(runtime_option)
|
||||||
@@ -34,40 +41,53 @@ class YOLOv6(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv6 initialize failed."
|
assert self.initialized, "YOLOv6 initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
def use_cuda_preprocessing(self, max_image_size=3840 * 2160):
|
|
||||||
return self._model.use_cuda_preprocessing(max_image_size)
|
|
||||||
|
|
||||||
# 一些跟YOLOv6模型有关的属性封装
|
# 一些跟YOLOv6模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -95,19 +115,21 @@ class YOLOv6(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@max_wh.setter
|
@max_wh.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv7(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv7 model exported by YOLOv7.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov7.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv7, self).__init__(runtime_option)
|
super(YOLOv7, self).__init__(runtime_option)
|
||||||
@@ -34,40 +41,53 @@ class YOLOv7(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv7 initialize failed."
|
assert self.initialized, "YOLOv7 initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
def use_cuda_preprocessing(self, max_image_size=3840 * 2160):
|
|
||||||
return self._model.use_cuda_preprocessing(max_image_size)
|
|
||||||
|
|
||||||
# 一些跟YOLOv7模型有关的属性封装
|
# 一些跟YOLOv7模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -95,19 +115,21 @@ class YOLOv7(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@max_wh.setter
|
@max_wh.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv7End2EndORT(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv7End2EndORT model exported by YOLOv7.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov7end2end_ort.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv7End2EndORT, self).__init__(runtime_option)
|
super(YOLOv7End2EndORT, self).__init__(runtime_option)
|
||||||
@@ -34,32 +41,46 @@ class YOLOv7End2EndORT(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv7End2End initialize failed."
|
assert self.initialized, "YOLOv7End2End initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25):
|
def predict(self, input_image, conf_threshold=0.25):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold)
|
return self._model.predict(input_image, conf_threshold)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -87,17 +108,19 @@ class YOLOv7End2EndORT(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv7End2EndTRT(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv7End2EndTRT model exported by YOLOv7.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov7end2end_trt.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv7End2EndTRT, self).__init__(runtime_option)
|
super(YOLOv7End2EndTRT, self).__init__(runtime_option)
|
||||||
@@ -34,35 +41,46 @@ class YOLOv7End2EndTRT(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv7End2EndTRT initialize failed."
|
assert self.initialized, "YOLOv7End2EndTRT initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25):
|
def predict(self, input_image, conf_threshold=0.25):
|
||||||
return self._model.predict(input_image, conf_threshold)
|
"""Detect an input image
|
||||||
|
|
||||||
def use_cuda_preprocessing(self, max_image_size=3840 * 2160):
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
return self._model.use_cuda_preprocessing(max_image_size)
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
|
return self._model.predict(input_image, conf_threshold)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -90,17 +108,19 @@ class YOLOv7End2EndTRT(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOX(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOX model exported by YOLOX.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolox.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOX, self).__init__(runtime_option)
|
super(YOLOX, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class YOLOX(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOX initialize failed."
|
assert self.initialized, "YOLOX initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: DetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,22 +55,35 @@ class YOLOX(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_decode_exported(self):
|
def is_decode_exported(self):
|
||||||
|
"""
|
||||||
|
whether the model_file was exported with decode module.
|
||||||
|
The official YOLOX/tools/export_onnx.py script will export ONNX file without decode module.
|
||||||
|
Please set it 'true' manually if the model file was exported with decode module.
|
||||||
|
"""
|
||||||
return self._model.is_decode_exported
|
return self._model.is_decode_exported
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def downsample_strides(self):
|
def downsample_strides(self):
|
||||||
|
"""
|
||||||
|
downsample strides for YOLOX to generate anchors, will take (8,16,32) as default values, might have stride=64.
|
||||||
|
"""
|
||||||
return self._model.downsample_strides
|
return self._model.downsample_strides
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def max_wh(self):
|
def max_wh(self):
|
||||||
|
# for offseting the boxes by classes when using NMS
|
||||||
return self._model.max_wh
|
return self._model.max_wh
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class RetinaFace(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a RetinaFace model exported by RetinaFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./retinaface.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(RetinaFace, self).__init__(runtime_option)
|
super(RetinaFace, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class RetinaFace(FastDeployModel):
|
|||||||
assert self.initialized, "RetinaFace initialize failed."
|
assert self.initialized, "RetinaFace initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.7, nms_iou_threshold=0.3):
|
def predict(self, input_image, conf_threshold=0.7, nms_iou_threshold=0.3):
|
||||||
|
"""Detect the location and key points of human faces from an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.7
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.3
|
||||||
|
:return: FaceDetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,22 +55,37 @@ class RetinaFace(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [640, 480]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [640, 480]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def variance(self):
|
def variance(self):
|
||||||
|
"""
|
||||||
|
Argument for image postprocessing step, variance in RetinaFace's prior-box(anchor) generate process, default (0.1, 0.2)
|
||||||
|
"""
|
||||||
return self._model.variance
|
return self._model.variance
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def downsample_strides(self):
|
def downsample_strides(self):
|
||||||
|
"""
|
||||||
|
Argument for image postprocessing step, downsample strides (namely, steps) for RetinaFace to generate anchors, will take (8,16,32) as default values
|
||||||
|
"""
|
||||||
return self._model.downsample_strides
|
return self._model.downsample_strides
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def min_sizes(self):
|
def min_sizes(self):
|
||||||
|
"""
|
||||||
|
Argument for image postprocessing step, min sizes, width and height for each anchor
|
||||||
|
"""
|
||||||
return self._model.min_sizes
|
return self._model.min_sizes
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def landmarks_per_face(self):
|
def landmarks_per_face(self):
|
||||||
|
"""
|
||||||
|
Argument for image postprocessing step, landmarks_per_face, default 5 in RetinaFace
|
||||||
|
"""
|
||||||
return self._model.landmarks_per_face
|
return self._model.landmarks_per_face
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,16 +24,30 @@ class SCRFD(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a SCRFD model exported by SCRFD.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./scrfd.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(SCRFD, self).__init__(runtime_option)
|
super(SCRFD, self).__init__(runtime_option)
|
||||||
|
|
||||||
self._model = C.vision.facedet.SCRFD(model_file, params_file,
|
self._model = C.vision.facedet.SCRFD(
|
||||||
self._runtime_option, model_format)
|
model_file, params_file, self._runtime_option, model_format)
|
||||||
# 通过self.initialized判断整个模型的初始化是否成功
|
# 通过self.initialized判断整个模型的初始化是否成功
|
||||||
assert self.initialized, "SCRFD initialize failed."
|
assert self.initialized, "SCRFD initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.7, nms_iou_threshold=0.3):
|
def predict(self, input_image, conf_threshold=0.7, nms_iou_threshold=0.3):
|
||||||
|
"""Detect the location and key points of human faces from an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.7
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.3
|
||||||
|
:return: FaceDetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,26 +55,34 @@ class SCRFD(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [640, 640]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [640, 640]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
@@ -108,19 +130,21 @@ class SCRFD(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@downsample_strides.setter
|
@downsample_strides.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class UltraFace(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a UltraFace model exported by UltraFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./ultraface.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(UltraFace, self).__init__(runtime_option)
|
super(UltraFace, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class UltraFace(FastDeployModel):
|
|||||||
assert self.initialized, "UltraFace initialize failed."
|
assert self.initialized, "UltraFace initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.7, nms_iou_threshold=0.3):
|
def predict(self, input_image, conf_threshold=0.7, nms_iou_threshold=0.3):
|
||||||
|
"""Detect the location and key points of human faces from an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.7
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.3
|
||||||
|
:return: FaceDetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,6 +55,9 @@ class UltraFace(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [640, 480]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [640, 480]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class YOLOv5Face(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a YOLOv5Face model exported by YOLOv5Face.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./yolov5face.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(YOLOv5Face, self).__init__(runtime_option)
|
super(YOLOv5Face, self).__init__(runtime_option)
|
||||||
@@ -34,6 +41,13 @@ class YOLOv5Face(FastDeployModel):
|
|||||||
assert self.initialized, "YOLOv5Face initialize failed."
|
assert self.initialized, "YOLOv5Face initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
def predict(self, input_image, conf_threshold=0.25, nms_iou_threshold=0.5):
|
||||||
|
"""Detect the location and key points of human faces from an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:param conf_threshold: confidence threashold for postprocessing, default is 0.25
|
||||||
|
:param nms_iou_threshold: iou threashold for NMS, default is 0.5
|
||||||
|
:return: FaceDetectionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image, conf_threshold,
|
return self._model.predict(input_image, conf_threshold,
|
||||||
nms_iou_threshold)
|
nms_iou_threshold)
|
||||||
|
|
||||||
@@ -41,30 +55,41 @@ class YOLOv5Face(FastDeployModel):
|
|||||||
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [1280, 1280]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def padding_value(self):
|
def padding_value(self):
|
||||||
|
# padding value, size should be the same as channels
|
||||||
return self._model.padding_value
|
return self._model.padding_value
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_no_pad(self):
|
def is_no_pad(self):
|
||||||
|
# while is_mini_pad = false and is_no_pad = true, will resize the image to the set size
|
||||||
return self._model.is_no_pad
|
return self._model.is_no_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_mini_pad(self):
|
def is_mini_pad(self):
|
||||||
|
# only pad to the minimum rectange which height and width is times of stride
|
||||||
return self._model.is_mini_pad
|
return self._model.is_mini_pad
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def is_scale_up(self):
|
def is_scale_up(self):
|
||||||
|
# if is_scale_up is false, the input image only can be zoom out, the maximum resize scale cannot exceed 1.0
|
||||||
return self._model.is_scale_up
|
return self._model.is_scale_up
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def stride(self):
|
def stride(self):
|
||||||
|
# padding stride, for is_mini_pad
|
||||||
return self._model.stride
|
return self._model.stride
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def landmarks_per_face(self):
|
def landmarks_per_face(self):
|
||||||
|
"""
|
||||||
|
Argument for image postprocessing step, landmarks_per_face, default 5 in YOLOv5Face
|
||||||
|
"""
|
||||||
return self._model.landmarks_per_face
|
return self._model.landmarks_per_face
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
@@ -92,19 +117,21 @@ class YOLOv5Face(FastDeployModel):
|
|||||||
@is_mini_pad.setter
|
@is_mini_pad.setter
|
||||||
def is_mini_pad(self, value):
|
def is_mini_pad(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_mini_pad` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_mini_pad` must be type of bool."
|
||||||
self._model.is_mini_pad = value
|
self._model.is_mini_pad = value
|
||||||
|
|
||||||
@is_scale_up.setter
|
@is_scale_up.setter
|
||||||
def is_scale_up(self, value):
|
def is_scale_up(self, value):
|
||||||
assert isinstance(
|
assert isinstance(
|
||||||
value, bool), "The value to set `is_scale_up` must be type of bool."
|
value,
|
||||||
|
bool), "The value to set `is_scale_up` must be type of bool."
|
||||||
self._model.is_scale_up = value
|
self._model.is_scale_up = value
|
||||||
|
|
||||||
@stride.setter
|
@stride.setter
|
||||||
def stride(self, value):
|
def stride(self, value):
|
||||||
assert isinstance(value,
|
assert isinstance(
|
||||||
int), "The value to set `stride` must be type of int."
|
value, int), "The value to set `stride` must be type of int."
|
||||||
self._model.stride = value
|
self._model.stride = value
|
||||||
|
|
||||||
@landmarks_per_face.setter
|
@landmarks_per_face.setter
|
||||||
|
|||||||
@@ -23,6 +23,13 @@ class AdaFace(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.PADDLE):
|
model_format=ModelFormat.PADDLE):
|
||||||
|
"""Load a AdaFace model exported by InsigtFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./adaface.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(AdaFace, self).__init__(runtime_option)
|
super(AdaFace, self).__init__(runtime_option)
|
||||||
@@ -33,28 +40,48 @@ class AdaFace(FastDeployModel):
|
|||||||
assert self.initialized, "AdaFace initialize failed."
|
assert self.initialized, "AdaFace initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the face recognition result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: FaceRecognitionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def l2_normalize(self):
|
def l2_normalize(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to apply l2 normalize to embedding values, default;
|
||||||
|
"""
|
||||||
return self._model.l2_normalize
|
return self._model.l2_normalize
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -25,6 +25,13 @@ class ArcFace(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a ArcFace model exported by InsigtFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./arcface.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(ArcFace, self).__init__(runtime_option)
|
super(ArcFace, self).__init__(runtime_option)
|
||||||
@@ -35,28 +42,48 @@ class ArcFace(FastDeployModel):
|
|||||||
assert self.initialized, "ArcFace initialize failed."
|
assert self.initialized, "ArcFace initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the face recognition result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: FaceRecognitionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def l2_normalize(self):
|
def l2_normalize(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to apply l2 normalize to embedding values, default;
|
||||||
|
"""
|
||||||
return self._model.l2_normalize
|
return self._model.l2_normalize
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class CosFace(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a CosFace model exported by InsigtFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./cosface.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(CosFace, self).__init__(runtime_option)
|
super(CosFace, self).__init__(runtime_option)
|
||||||
@@ -34,28 +41,48 @@ class CosFace(FastDeployModel):
|
|||||||
assert self.initialized, "CosFace initialize failed."
|
assert self.initialized, "CosFace initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the face recognition result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: FaceRecognitionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def l2_normalize(self):
|
def l2_normalize(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to apply l2 normalize to embedding values, default False;
|
||||||
|
"""
|
||||||
return self._model.l2_normalize
|
return self._model.l2_normalize
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class InsightFaceRecognitionModel(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a InsightFace model exported by InsigtFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./arcface.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(InsightFaceRecognitionModel, self).__init__(runtime_option)
|
super(InsightFaceRecognitionModel, self).__init__(runtime_option)
|
||||||
@@ -34,28 +41,48 @@ class InsightFaceRecognitionModel(FastDeployModel):
|
|||||||
assert self.initialized, "InsightFaceRecognitionModel initialize failed."
|
assert self.initialized, "InsightFaceRecognitionModel initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the face recognition result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: FaceRecognitionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟InsightFaceRecognitionModel模型有关的属性封装
|
# 一些跟InsightFaceRecognitionModel模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def l2_normalize(self):
|
def l2_normalize(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to apply l2 normalize to embedding values, default False;
|
||||||
|
"""
|
||||||
return self._model.l2_normalize
|
return self._model.l2_normalize
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class PartialFC(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a PartialFC model exported by InsigtFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./partial_fc.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(PartialFC, self).__init__(runtime_option)
|
super(PartialFC, self).__init__(runtime_option)
|
||||||
@@ -34,28 +41,48 @@ class PartialFC(FastDeployModel):
|
|||||||
assert self.initialized, "PartialFC initialize failed."
|
assert self.initialized, "PartialFC initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the face recognition result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: FaceRecognitionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def l2_normalize(self):
|
def l2_normalize(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to apply l2 normalize to embedding values, default False;
|
||||||
|
"""
|
||||||
return self._model.l2_normalize
|
return self._model.l2_normalize
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class VPL(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a VPL model exported by InsigtFace.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./vpl.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(VPL, self).__init__(runtime_option)
|
super(VPL, self).__init__(runtime_option)
|
||||||
@@ -34,28 +41,48 @@ class VPL(FastDeployModel):
|
|||||||
assert self.initialized, "VPL initialize failed."
|
assert self.initialized, "VPL initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the face recognition result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: FaceRecognitionResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [112, 112]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def l2_normalize(self):
|
def l2_normalize(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to apply l2 normalize to embedding values, default False;
|
||||||
|
"""
|
||||||
return self._model.l2_normalize
|
return self._model.l2_normalize
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class MODNet(FastDeployModel):
|
|||||||
params_file="",
|
params_file="",
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.ONNX):
|
model_format=ModelFormat.ONNX):
|
||||||
|
"""Load a MODNet model exported by MODNet.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g ./modnet.onnx
|
||||||
|
:param params_file: (str)Path of parameters file, e.g yolox/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
# 调用基函数进行backend_option的初始化
|
# 调用基函数进行backend_option的初始化
|
||||||
# 初始化后的option保存在self._runtime_option
|
# 初始化后的option保存在self._runtime_option
|
||||||
super(MODNet, self).__init__(runtime_option)
|
super(MODNet, self).__init__(runtime_option)
|
||||||
@@ -34,24 +41,41 @@ class MODNet(FastDeployModel):
|
|||||||
assert self.initialized, "MODNet initialize failed."
|
assert self.initialized, "MODNet initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the matting result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: MattingResult
|
||||||
|
"""
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|
||||||
# 一些跟模型有关的属性封装
|
# 一些跟模型有关的属性封装
|
||||||
# 多数是预处理相关，可通过修改如model.size = [256, 256]改变预处理时resize的大小（前提是模型支持）
|
# 多数是预处理相关，可通过修改如model.size = [256, 256]改变预处理时resize的大小（前提是模型支持）
|
||||||
@property
|
@property
|
||||||
def size(self):
|
def size(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, the preprocess image size, tuple of (width, height)
|
||||||
|
"""
|
||||||
return self._model.size
|
return self._model.size
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def alpha(self):
|
def alpha(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, alpha value for normalization
|
||||||
|
"""
|
||||||
return self._model.alpha
|
return self._model.alpha
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def beta(self):
|
def beta(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, beta value for normalization
|
||||||
|
"""
|
||||||
return self._model.beta
|
return self._model.beta
|
||||||
|
|
||||||
@property
|
@property
|
||||||
def swap_rb(self):
|
def swap_rb(self):
|
||||||
|
"""
|
||||||
|
Argument for image preprocessing step, whether to swap the B and R channel, such as BGR->RGB, default true.
|
||||||
|
"""
|
||||||
return self._model.swap_rb
|
return self._model.swap_rb
|
||||||
|
|
||||||
@size.setter
|
@size.setter
|
||||||
|
|||||||
@@ -25,6 +25,14 @@ class PPMatting(FastDeployModel):
|
|||||||
config_file,
|
config_file,
|
||||||
runtime_option=None,
|
runtime_option=None,
|
||||||
model_format=ModelFormat.PADDLE):
|
model_format=ModelFormat.PADDLE):
|
||||||
|
"""Load a PPMatting model exported by PaddleSeg.
|
||||||
|
|
||||||
|
:param model_file: (str)Path of model file, e.g PPMatting-512/model.pdmodel
|
||||||
|
:param params_file: (str)Path of parameters file, e.g PPMatting-512/model.pdiparams, if the model_fomat is ModelFormat.ONNX, this param will be ignored, can be set as empty string
|
||||||
|
:param config_file: (str)Path of configuration file for deployment, e.g PPMatting-512/deploy.yml
|
||||||
|
:param runtime_option: (fastdeploy.RuntimeOption)RuntimeOption for inference this model, if it's None, will use the default backend on CPU
|
||||||
|
:param model_format: (fastdeploy.ModelForamt)Model format of the loaded model
|
||||||
|
"""
|
||||||
super(PPMatting, self).__init__(runtime_option)
|
super(PPMatting, self).__init__(runtime_option)
|
||||||
|
|
||||||
assert model_format == ModelFormat.PADDLE, "PPMatting model only support model format of ModelFormat.Paddle now."
|
assert model_format == ModelFormat.PADDLE, "PPMatting model only support model format of ModelFormat.Paddle now."
|
||||||
@@ -34,5 +42,10 @@ class PPMatting(FastDeployModel):
|
|||||||
assert self.initialized, "PPMatting model initialize failed."
|
assert self.initialized, "PPMatting model initialize failed."
|
||||||
|
|
||||||
def predict(self, input_image):
|
def predict(self, input_image):
|
||||||
|
""" Predict the matting result for an input image
|
||||||
|
|
||||||
|
:param input_image: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format
|
||||||
|
:return: MattingResult
|
||||||
|
"""
|
||||||
assert input_image is not None, "The input image data is None."
|
assert input_image is not None, "The input image data is None."
|
||||||
return self._model.predict(input_image)
|
return self._model.predict(input_image)
|
||||||
|
|||||||
Reference in New Issue
Block a user