Rebuilting the CMU-OpenPose pose estimatior using Python with OpenCV and Tensorflow. (The code comments are partly descibed in chinese)
Pretrained-model Downloading
In this work, I used both caffemodel and tensorflow-graph-model, you can download them here, Then place the pretrained models to corresponding directory respectively.
Examples:
place caffe_modelsposebody_25pose_iter_584000.caffemodel into pose-estimator-using-caffemodelmodelbody_25
place caffe_modelshandpose_iter_102000.caffemodel into hand-estimator-using-caffemodelmodel
place openpose graph model cocograph_opt.pb into pose-estimator-tensorflowgraph_model_coco
Requirements :
OpenCV > 3.4.1
TensorFlow > 1.2.0
imutils
Usage:
See the sub-README.md in sub-folder.
BODY_25 vs. COCO vs. MPI
BODY_25 model is faster, more accurate, and it includes foot keypoints.
COCO requires less memory on GPU (being able to fit into 2GB GPUs with the default settings) and it runs faster on CPU-only mode.
MPI model is only meant for people requiring the MPI-keypoint structure. It is also slower than BODY_25 and far less accurate.
Output Format
Body_25 in left, COCO in middle, MPI in right.
See more Output Format details here, and Hand Output Format included as well.