Skip to content

Latest commit

 

History

History

cpp

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
CMakeLists.txt
CMakeLists.txt
 
 
README.md
README.md
 
 
README_CN.md
README_CN.md
 
 
det_keypoint_unite_infer.cc
det_keypoint_unite_infer.cc
 
 

README.md

English | 简体中文

PP-PicoDet + PP-TinyPose (Pipeline) C++ Deployment Example

This directory provides the Multi-person keypoint detection in a single image example that det_keypoint_unite_infer.cc fast finishes the deployment of multi-person detection model PP-PicoDet + PP-TinyPose on CPU/GPU and GPU accelerated by TensorRT. The script is as follows

Before deployment, two steps require confirmation

Taking the inference on Linux as an example, the compilation test can be completed by executing the following command in this directory. FastDeploy version 0.7.0 or above (x.x.x>=0.7.0) is required to support this model.

mkdir build
cd build
# Download the FastDeploy precompiled library. Users can choose your appropriate version in the `FastDeploy  Precompiled Library` mentioned above 
wget https://bj.bcebos.com/fastdeploy/release/cpp/fastdeploy-linux-x64-x.x.x.tgz
tar xvf fastdeploy-linux-x64-x.x.x.tgz
cmake .. -DFASTDEPLOY_INSTALL_DIR=${PWD}/fastdeploy-linux-x64-x.x.x
make -j

# Download PP-TinyPose+PP-PicoDet 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/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 inference
./infer_demo PP_PicoDet_V2_S_Pedestrian_320x320_infer PP_TinyPose_256x192_infer 000000018491.jpg 0
# GPU inference
./infer_demo PP_PicoDet_V2_S_Pedestrian_320x320_infer PP_TinyPose_256x192_infer 000000018491.jpg 1
# TensorRT inference on GPU
./infer_demo PP_PicoDet_V2_S_Pedestrian_320x320_infer PP_TinyPose_256x192_infer 000000018491.jpg 2
# kunlunxin XPU inference
./infer_demo PP_PicoDet_V2_S_Pedestrian_320x320_infer PP_TinyPose_256x192_infer 000000018491.jpg 3

The visualized result after running is as follows

The above command works for Linux or MacOS. For SDK use-pattern in Windows, refer to:

PP-TinyPose C++ Interface

PP-TinyPose Class

fastdeploy::pipeline::PPTinyPose(
        fastdeploy::vision::detection::PPYOLOE* det_model,
        fastdeploy::vision::keypointdetection::PPTinyPose* pptinypose_model)

PPTinyPose Pipeline model loading and initialization.

Parameter

  • model_det_modelfile(fastdeploy::vision::detection): Initialized detection model. Refer to PP-TinyPose
  • pptinypose_model(fastdeploy::vision::keypointdetection): Initialized detection model Detection. Currently only PaddleDetection series is available.

Predict Function

PPTinyPose::Predict(cv::Mat* im, KeyPointDetectionResult* result)

Model prediction interface. Input images and output keypoint detection results.

Parameter

  • im: Input images in HWC or BGR format
  • result: Keypoint detection results, including coordinates and the corresponding probability value. Refer to Vision Model Prediction Results for the description of KeyPointDetectionResult

Class Member Property

Post-processing Parameter

  • detection_model_score_threshold(bool): Score threshold of the Detectin model for filtering detection boxes before entering the PP-TinyPose model