From e811e9e239c85f18a4a74afa9c72356f6823722c Mon Sep 17 00:00:00 2001 From: ziqi-jin <67993288+ziqi-jin@users.noreply.github.com> Date: Thu, 1 Dec 2022 20:46:35 +0800 Subject: [PATCH] [Doc] API docs for Visualize module (#770) * 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 * 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 Co-authored-by: root 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 * 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 Co-authored-by: root 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 * 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 Co-authored-by: root Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason Co-authored-by: root 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 * 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 Co-authored-by: root 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 * 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 Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason Co-authored-by: root 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 * comments for vis * python API * python API Co-authored-by: Jason Co-authored-by: root Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com> Co-authored-by: huangjianhui <852142024@qq.com> Co-authored-by: Jason <928090362@qq.com> --- docs/api_docs/cpp/Doxyfile | 2 +- docs/api_docs/python/visualize.md | 57 +++++++++++++ fastdeploy/vision/visualize/visualize.h | 84 ++++++++++++++++++- .../fastdeploy/vision/visualize/__init__.py | 52 ++++++++++++ 4 files changed, 193 insertions(+), 2 deletions(-) create mode 100644 docs/api_docs/python/visualize.md diff --git a/docs/api_docs/cpp/Doxyfile b/docs/api_docs/cpp/Doxyfile index 4bd48f09d..afd956162 100644 --- a/docs/api_docs/cpp/Doxyfile +++ b/docs/api_docs/cpp/Doxyfile @@ -2100,7 +2100,7 @@ INCLUDE_FILE_PATTERNS = # recursively expanded use the := operator instead of the = operator. # This tag requires that the tag ENABLE_PREPROCESSING is set to YES. -PREDEFINED = protected=private +PREDEFINED = protected=private ENABLE_VISION_VISUALIZE=1 # If the MACRO_EXPANSION and EXPAND_ONLY_PREDEF tags are set to YES then this # tag can be used to specify a list of macro names that should be expanded. The diff --git a/docs/api_docs/python/visualize.md b/docs/api_docs/python/visualize.md new file mode 100644 index 000000000..2170eb453 --- /dev/null +++ b/docs/api_docs/python/visualize.md @@ -0,0 +1,57 @@ +# Visaulize(可视化) + +## fastdeploy.vision.vis_detection + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_detection + :members: + :inherited-members: +``` + +## fastdeploy.vision.vis_segmentation + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_segmentation + :members: + :inherited-members: +``` + +## fastdeploy.vision.vis_keypoint_detection + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_keypoint_detection + :members: + :inherited-members: +``` +## fastdeploy.vision.vis_face_detection + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_face_detection + :members: + :inherited-members: +``` + + +## fastdeploy.vision.vis_face_alignment + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_face_alignment + :members: + :inherited-members: +``` + +## fastdeploy.vision.vis_matting + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_matting + :members: + :inherited-members: +``` + +## fastdeploy.vision.vis_ppocr + +```{eval-rst} +.. autoclass:: fastdeploy.vision.vis_ppocr + :members: + :inherited-members: +``` diff --git a/fastdeploy/vision/visualize/visualize.h b/fastdeploy/vision/visualize/visualize.h index ea543553f..498944c23 100755 --- a/fastdeploy/vision/visualize/visualize.h +++ b/fastdeploy/vision/visualize/visualize.h @@ -20,9 +20,11 @@ #include "fastdeploy/vision/tracking/pptracking/model.h" namespace fastdeploy { +/** \brief All C++ FastDeploy Vision Models APIs are defined inside this namespace +* +*/ namespace vision { -// This class will deprecated, please not use it class FASTDEPLOY_DECL Visualize { public: static int num_classes_; @@ -52,35 +54,108 @@ class FASTDEPLOY_DECL Visualize { std::vector GenerateColorMap(int num_classes = 1000); cv::Mat RemoveSmallConnectedArea(const cv::Mat& alpha_pred, float threshold); +/** \brief Show the visualized results for detection models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] score_threshold threshold for result scores, the bounding box will not be shown if the score is less than score_threshold + * \param[in] line_size line size for bounding boxes + * \param[in] font_size font size for text + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisDetection(const cv::Mat& im, const DetectionResult& result, float score_threshold = 0.0, int line_size = 1, float font_size = 0.5f); +/** \brief Show the visualized results with custom labels for detection models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] labels the visualized result will show the bounding box contain class label + * \param[in] score_threshold threshold for result scores, the bounding box will not be shown if the score is less than score_threshold + * \param[in] line_size line size for bounding boxes + * \param[in] font_size font size for text + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisDetection(const cv::Mat& im, const DetectionResult& result, const std::vector& labels, float score_threshold = 0.0, int line_size = 1, float font_size = 0.5f); +/** \brief Show the visualized results for classification models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] top_k 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] score_threshold threshold for top_k scores, the class will not be shown if the score is less than score_threshold + * \param[in] font_size font size + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisClassification( const cv::Mat& im, const ClassifyResult& result, int top_k = 5, float score_threshold = 0.0f, float font_size = 0.5f); +/** \brief Show the visualized results with custom labels for classification models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] labels custom labels for user, the visualized result will show the corresponding custom labels + * \param[in] top_k 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] score_threshold threshold for top_k scores, the class will not be shown if the score is less than score_threshold + * \param[in] font_size font size + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisClassification( const cv::Mat& im, const ClassifyResult& result, const std::vector& labels, int top_k = 5, float score_threshold = 0.0f, float font_size = 0.5f); +/** \brief Show the visualized results for face detection models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] line_size line size for bounding boxes + * \param[in] font_size font size for text + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisFaceDetection(const cv::Mat& im, const FaceDetectionResult& result, int line_size = 1, float font_size = 0.5f); +/** \brief Show the visualized results for face alignment models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] line_size line size for circle point + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisFaceAlignment(const cv::Mat& im, const FaceAlignmentResult& result, int line_size = 1); +/** \brief Show the visualized results for segmentation models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] weight transparent weight of visualized result image + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisSegmentation(const cv::Mat& im, const SegmentationResult& result, float weight = 0.5); +/** \brief Show the visualized results for matting models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \param[in] remove_small_connected_area if remove_small_connected_area==true, the visualized result will not include the small connected areas + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisMatting(const cv::Mat& im, const MattingResult& result, bool remove_small_connected_area = false); +/** \brief Show the visualized results for Ocr models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] result the result produced by model + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisOcr(const cv::Mat& im, const OCRResult& ocr_result); FASTDEPLOY_DECL cv::Mat VisMOT(const cv::Mat& img, const MOTResult& results, @@ -93,6 +168,13 @@ FASTDEPLOY_DECL cv::Mat SwapBackground(const cv::Mat& im, const cv::Mat& background, const SegmentationResult& result, int background_label); +/** \brief Show the visualized results for key point detection models + * + * \param[in] im the input image data, comes from cv::imread(), is a 3-D array with layout HWC, BGR format + * \param[in] results the result produced by model + * \param[in] conf_threshold threshold for result scores, the result will not be shown if the score is less than conf_threshold + * \return cv::Mat type stores the visualized results + */ FASTDEPLOY_DECL cv::Mat VisKeypointDetection(const cv::Mat& im, const KeyPointDetectionResult& results, float conf_threshold = 0.5f); diff --git a/python/fastdeploy/vision/visualize/__init__.py b/python/fastdeploy/vision/visualize/__init__.py index ac640bf27..6ebd08c1c 100755 --- a/python/fastdeploy/vision/visualize/__init__.py +++ b/python/fastdeploy/vision/visualize/__init__.py @@ -24,25 +24,64 @@ def vis_detection(im_data, score_threshold=0.0, line_size=1, font_size=0.5): + """Show the visualized results for detection models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param det_result: the result produced by model + :param labels: (list of str) the visualized result will show the bounding box contain class label + :param score_threshold: (float) score_threshold threshold for result scores, the bounding box will not be shown if the score is less than score_threshold + :param line_size: (float) line_size line size for bounding boxes + :param font_size: (float) font_size font size for text + :return: (numpy.ndarray) image with visualized results + """ return C.vision.vis_detection(im_data, det_result, labels, score_threshold, line_size, font_size) def vis_keypoint_detection(im_data, keypoint_det_result, conf_threshold=0.5): + """Show the visualized results for keypoint detection models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param keypoint_det_result: the result produced by model + :param conf_threshold: (float) conf_threshold threshold for result scores, the bounding box will not be shown if the score is less than conf_threshold + :return: (numpy.ndarray) image with visualized results + """ return C.vision.Visualize.vis_keypoint_detection( im_data, keypoint_det_result, conf_threshold) def vis_face_detection(im_data, face_det_result, line_size=1, font_size=0.5): + """Show the visualized results for face detection models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param face_det_result: the result produced by model + :param line_size: (float) line_size line size for bounding boxes + :param font_size: (float) font_size font size for text + :return: (numpy.ndarray) image with visualized results + """ return C.vision.vis_face_detection(im_data, face_det_result, line_size, font_size) def vis_face_alignment(im_data, face_align_result, line_size=1): + """Show the visualized results for face alignment models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param face_align_result: the result produced by model + :param line_size: (float)line_size line size for circle point + :return: (numpy.ndarray) image with visualized results + """ return C.vision.vis_face_alignment(im_data, face_align_result, line_size) def vis_segmentation(im_data, seg_result, weight=0.5): + """Show the visualized results for segmentation models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param seg_result: the result produced by model + :param weight: (float)transparent weight of visualized result image + :return: (numpy.ndarray) image with visualized results + """ return C.vision.vis_segmentation(im_data, seg_result, weight) @@ -57,6 +96,13 @@ def vis_matting_alpha(im_data, def vis_matting(im_data, matting_result, remove_small_connected_area=False): + """Show the visualized results for matting models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param matting_result: the result produced by model + :param remove_small_connected_area: (bool) if remove_small_connected_area==True, the visualized result will not include the small connected areas + :return: (numpy.ndarray) image with visualized results + """ return C.vision.vis_matting(im_data, matting_result, remove_small_connected_area) @@ -105,6 +151,12 @@ def swap_background(im_data, def vis_ppocr(im_data, det_result): + """Show the visualized results for ocr models + + :param im_data: (numpy.ndarray)The input image data, 3-D array with layout HWC, BGR format + :param det_result: the result produced by model + :return: (numpy.ndarray) image with visualized results + """ return C.vision.vis_ppocr(im_data, det_result)