Open chenyu19880302 opened 3 years ago
While, the original paper writes "The tracker predicts keypoint coordinates, the presence of the human on the current frame, and the refined region of interest for the current frame". On my personal note, I think this algorithm is to detect the joint of the head (i.e. BlazeFace) and the middle point of his or her hips. Then, these two points are utilized as a ruler to calculate the offsets of the position and angle of the person. Afterwards, other keypoints are detected according to the information of the torso.
Hello!Sorry to interrupt you. I am comfused about the tracker part on blazepose. I just understand the whole process is that detect the first frame,then tracker the next frame according to the pre-frame's skeleton position.But I don't understand the tracker part on blazepose.Can you explain the tracker part on blazepose? Thank you very much!