English | [简体中文](README_CN.md) # PP-TinyPose Python Deployment Example Before deployment, two steps require confirmation - 1. Software and hardware should meet the requirements. Please refer to [FastDeploy Environment Requirements](../../../../../docs/en/build_and_install/download_prebuilt_libraries.md) - 2. Download the precompiled deployment library and samples code according to your development environment. Refer to [FastDeploy Precompiled Library](../../../../../docs/en/build_and_install/download_prebuilt_libraries.md) This directory provides the `Multi-person keypoint detection in a single image` example that `pptinypose_infer.py` fast finishes the deployment of PP-TinyPose on CPU/GPU and GPU accelerated by TensorRT. The script is as follows >> **Attention**: single model currently only supports single-person keypoint detection in a single image. Therefore, the input image should contain one person only or should be cropped. For multi-person keypoint detection, refer to [PP-TinyPose Pipeline](../../det_keypoint_unite/python/README.md) ```bash # Download the example code for deployment git clone https://github.com/PaddlePaddle/FastDeploy.git cd FastDeploy/examples/vision/keypointdetection/tiny_pose/python # Download PP-TinyPose model files and test images 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/hrnet_demo.jpg # CPU inference python pptinypose_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --image hrnet_demo.jpg --device cpu # GPU inference python pptinypose_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --image hrnet_demo.jpg --device gpu # TensorRT inference on GPU(Attention: It is somewhat time-consuming for the operation of model serialization when running TensorRT inference for the first time. Please be patient.) python pptinypose_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --image hrnet_demo.jpg --device gpu --use_trt True # KunlunXin XPU inference python pptinypose_infer.py --tinypose_model_dir PP_TinyPose_256x192_infer --image hrnet_demo.jpg --device kunlunxin ``` The visualized result after running is as follows
## PP-TinyPose Python Interface ```python fd.vision.keypointdetection.PPTinyPose(model_file, params_file, config_file, runtime_option=None, model_format=ModelFormat.PADDLE) ``` PP-TinyPose model loading and initialization, among which model_file, params_file, and config_file are the Paddle inference files exported from the training model. Refer to [Model Export](https://github.com/PaddlePaddle/PaddleDetection/blob/release/2.5/deploy/EXPORT_MODEL.md) for more information **Parameter** > * **model_file**(str): Model file path > * **params_file**(str): Parameter file path > * **config_file**(str): Inference deployment configuration file > * **runtime_option**(RuntimeOption): Backend inference configuration. None by default, which is the default configuration > * **model_format**(ModelFormat): Model format. Paddle format by default ### predict function > ```python > PPTinyPose.predict(input_image) > ``` > > Model prediction interface. Input images and output detection results. > > **Parameter** > > > * **input_image**(np.ndarray): Input data in HWC or BGR format > **Return** > > > Return `fastdeploy.vision.KeyPointDetectionResult` structure. Refer to [Vision Model Prediction Results](../../../../../docs/api/vision_results/) for the description of the structure. ### Class Member Property #### Post-processing Parameter Users can modify the following pre-processing parameters to their needs, which affects the final inference and deployment results > > * **use_dark**(bool): • Whether to use DARK for post-processing. Refer to [Reference Paper](https://arxiv.org/abs/1910.06278) ## Other Documents - [PP-TinyPose Model Description](..) - [PP-TinyPose C++ Deployment](../cpp) - [Model Prediction Results](../../../../../docs/api/vision_results/) - [How to switch the model inference backend engine](../../../../../docs/en/faq/how_to_change_backend.md)