[Model] Add tinypose single && pipeline model (#177)

* Add tinypose model * Add PPTinypose python API * Fix picodet preprocess bug && Add Tinypose examples * Update tinypose example code * Update ppseg preprocess if condition * Update ppseg backend support type * Update permute.h * Update README.md * Update code with comments * Move files dir * Delete premute.cc * Add single model pptinypose * Delete pptinypose old code in ppdet * Code format * Add ppdet + pptinypose pipeline model * Fix bug for posedetpipeline * Change Frontend to ModelFormat * Change Frontend to ModelFormat in __init__.py * Add python posedetpipeline/ * Update pptinypose example dir name * Update README.md * Update README.md * Update README.md * Update README.md * Create keypointdetection_result.md * Create README.md * Create README.md * Create README.md * Update README.md * Update README.md * Create README.md * Fix det_keypoint_unite_infer.py bug * Create README.md * Update PP-Tinypose by comment * Update by comment * Add pipeline directory * Add pptinypose dir * Update pptinypose to align accuracy * Addd warpAffine processor * Update GetCpuMat to GetOpenCVMat * Add comment for pptinypose && pipline * Update docs/main_page.md * Add README.md for pptinypose * Add README for det_keypoint_unite * Remove ENABLE_PIPELINE option * Remove ENABLE_PIPELINE option * Change pptinypose default backend * PP-TinyPose Pipeline support multi PP-Detection models * Update pp-tinypose comment * Update by comments * Add single test example Co-authored-by: Jason <jiangjiajun@baidu.com>
2025-10-05 16:48:03 +08:00 · 2022-10-21 09:28:23 +08:00
parent 49ab773d22
commit b565c15bf7
62 changed files with 2583 additions and 20 deletions
--- a/examples/vision/keypointdetection/det_keypoint_unite/python/README.md
+++ b/examples/vision/keypointdetection/det_keypoint_unite/python/README.md
@@ -0,0 +1,74 @@
+# PP-PicoDet + PP-TinyPose (Pipeline) Python部署示例
+
+在部署前，需确认以下两个步骤
+
+- 1. 软硬件环境满足要求，参考[FastDeploy环境要求](../../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)  
+- 2. 根据开发环境，下载预编译部署库和samples代码，参考[FastDeploy预编译库](../../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
+
+本目录下提供`det_keypoint_unite_infer.py`快速完成多人模型配置 PP-PicoDet + PP-TinyPose 在CPU/GPU，以及GPU上通过TensorRT加速部署的`单图多人关键点检测`示例。执行如下脚本即可完成
+>> **注意**: PP-TinyPose单模型独立部署，请参考[PP-TinyPose 单模型](../../tiny_pose//python/README.md)
+
+```bash
+#下载部署示例代码
+git clone https://github.com/PaddlePaddle/FastDeploy.git
+cd FastDeploy/examples/vision/keypointdetection/det_keypoint_unite/python
+
+# 下载PP-TinyPose模型文件和测试图片
+wget https://bj.bcebos.com/paddlehub/fastdeploy/PP_TinyPose_256x192_infer.tgz
+tar -xvf PP_TinyPose_256x192_infer.tgz
+wget https://bj.bcebos.com/paddlehub/fastdeploy/PP_PicoDet_V2_S_Pedestrian_320x320_infer.tgz
+tar -xvf PP_PicoDet_V2_S_Pedestrian_320x320_infer.tgz
+wget https://bj.bcebos.com/paddlehub/fastdeploy/000000018491.jpg
+# CPU推理
+python det_keypoint_unite_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --det_model_dir PP_PicoDet_V2_S_Pedestrian_320x320_infer --image 000000018491.jpg --device cpu
+# GPU推理
+python det_keypoint_unite_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --det_model_dir PP_PicoDet_V2_S_Pedestrian_320x320_infer --image 000000018491.jpg --device gpu
+# GPU上使用TensorRT推理 （注意：TensorRT推理第一次运行，有序列化模型的操作，有一定耗时，需要耐心等待）
+python det_keypoint_unite_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --det_model_dir PP_PicoDet_V2_S_Pedestrian_320x320_infer --image 000000018491.jpg --device gpu --use_trt True
+```
+
+运行完成可视化结果如下图所示
+<div  align="center">  
+<img src="https://user-images.githubusercontent.com/16222477/196393343-eeb6b68f-0bc6-4927-871f-5ac610da7293.jpeg", width=640px, height=427px />
+</div>
+
+## PPTinyPosePipeline Python接口
+
+```python
+fd.pipeline.PPTinyPose(det_model=None, pptinypose_model=None)
+```
+
+PPTinyPosePipeline模型加载和初始化，其中det_model是使用`fd.vision.detection.PicoDet`[参考Detection文档](../../../detection/paddledetection/python/)初始化的检测模型，pptinypose_model是使用`fd.vision.keypointdetection.PPTinyPose`[参考PP-TinyPose文档](../../tiny_pose/python/)初始化的检测模型
+
+**参数**
+
+> * **det_model**(str): 初始化后的检测模型
+> * **pptinypose_model**(str): 初始化后的PP-TinyPose模型
+
+### predict函数
+
+> ```python
+> PPTinyPosePipeline.predict(input_image)
+> ```
+>
+> 模型预测结口，输入图像直接输出检测结果。
+>
+> **参数**
+>
+> > * **input_image**(np.ndarray): 输入数据，注意需为HWC，BGR格式
+
+> **返回**
+>
+> > 返回`fastdeploy.vision.KeyPointDetectionResult`结构体，结构体说明参考文档[视觉模型预测结果](../../../../../docs/api/vision_results/)
+
+### 类成员属性
+#### 后处理参数
+> > * **detection_model_score_threshold**(bool):
+输入PP-TinyPose模型前，Detectin模型过滤检测框的分数阈值
+
+## 其它文档
+
+- [Pipeline 模型介绍](..)
+- [Pipeline C++部署](../cpp)
+- [模型预测结果说明](../../../../../docs/api/vision_results/)
+- [如何切换模型推理后端引擎](../../../../../docs/runtime/how_to_change_backend.md)
--- a/examples/vision/keypointdetection/det_keypoint_unite/python/det_keypoint_unite_infer.py
+++ b/examples/vision/keypointdetection/det_keypoint_unite/python/det_keypoint_unite_infer.py
@@ -0,0 +1,91 @@
+import fastdeploy as fd
+import cv2
+import os
+
+
+def parse_arguments():
+    import argparse
+    import ast
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--tinypose_model_dir",
+        required=True,
+        help="path of paddletinypose model directory")
+    parser.add_argument(
+        "--det_model_dir", help="path of paddledetection model directory")
+    parser.add_argument(
+        "--image", required=True, help="path of test image file.")
+    parser.add_argument(
+        "--device",
+        type=str,
+        default='cpu',
+        help="type of inference device, support 'cpu' or 'gpu'.")
+    parser.add_argument(
+        "--use_trt",
+        type=ast.literal_eval,
+        default=False,
+        help="wether to use tensorrt.")
+    return parser.parse_args()
+
+
+def build_picodet_option(args):
+    option = fd.RuntimeOption()
+
+    if args.device.lower() == "gpu":
+        option.use_gpu()
+
+    if args.use_trt:
+        option.use_trt_backend()
+        option.set_trt_input_shape("image", [1, 3, 320, 320])
+        option.set_trt_input_shape("scale_factor", [1, 2])
+    return option
+
+
+def build_tinypose_option(args):
+    option = fd.RuntimeOption()
+
+    if args.device.lower() == "gpu":
+        option.use_gpu()
+
+    if args.use_trt:
+        option.use_trt_backend()
+        option.set_trt_input_shape("image", [1, 3, 256, 192])
+    return option
+
+
+args = parse_arguments()
+picodet_model_file = os.path.join(args.det_model_dir, "model.pdmodel")
+picodet_params_file = os.path.join(args.det_model_dir, "model.pdiparams")
+picodet_config_file = os.path.join(args.det_model_dir, "infer_cfg.yml")
+
+# 配置runtime，加载模型
+runtime_option = build_picodet_option(args)
+det_model = fd.vision.detection.PicoDet(
+    picodet_model_file,
+    picodet_params_file,
+    picodet_config_file,
+    runtime_option=runtime_option)
+
+tinypose_model_file = os.path.join(args.tinypose_model_dir, "model.pdmodel")
+tinypose_params_file = os.path.join(args.tinypose_model_dir, "model.pdiparams")
+tinypose_config_file = os.path.join(args.tinypose_model_dir, "infer_cfg.yml")
+# 配置runtime，加载模型
+runtime_option = build_tinypose_option(args)
+tinypose_model = fd.vision.keypointdetection.PPTinyPose(
+    tinypose_model_file,
+    tinypose_params_file,
+    tinypose_config_file,
+    runtime_option=runtime_option)
+
+# 预测图片检测结果
+im = cv2.imread(args.image)
+pipeline = fd.pipeline.PPTinyPose(det_model, tinypose_model)
+pipeline.detection_model_score_threshold = 0.5
+pipeline_result = pipeline.predict(im)
+print("Paddle TinyPose Result:\n", pipeline_result)
+
+# 预测结果可视化
+vis_im = fd.vision.vis_keypoint_detection(
+    im, pipeline_result, conf_threshold=0.2)
+cv2.imwrite("visualized_result.jpg", vis_im)
+print("TinyPose visualized result save in ./visualized_result.jpg")