crossview_3d_pose_tracking
crossview_3d_pose_tracking copied to clipboard
longcw
Dataset of the paper "Cross-View Tracking for Multi-Human 3D Pose Estimation at over 100 FPS"
Thanks for your great work, may I ask if your speed evaluation includes the 2D pose estimation part? The paper mentioned that "our solution achieves 154 FPS on 12 cameras...
Hi, I found you refer to paper "Tracking Multiple People Online and in Real Time", which use "AL-ICM" method to solve correlation clustering problem, do you have the software which...
Hi, In your released code, 3D pose evaluation code is given. However, there is some question about tracking evaluation. In your paper, you mentioned that you projected 3d pose back...
Dear longcw there is visualizer.update(frame, frame_data) and visualizer.update_3d(data_3d) if I comment out visualizer.update_3d(data_3d) by #. it seems no difference with before. so excuse me how can just see the effect...
in https://openaccess.thecvf.com/content_CVPR_2020/papers/Chen_Cross-View_Tracking_for_Multi-Human_3D_Pose_Estimation_at_Over_100_CVPR_2020_paper.pdf formula 3 my code is: #P = K[T|R], P+ is pseudo-inverse of P def pseudo(points_2d, K, Tw): K =np.array(K) R,T = get_R_T(Tw) P = K.dot(np.hstack((R,T))) import scipy.linalg as...
Dear longcw thank your great job now I want to use back_project function to realize a single point in the image plane correspods to a ray in 3D space in...
Hi there~ Thank you for sharing this great work! My question is that how to use our own dataset to get 3D Human Pose Estimation with your pretrained model. And...
The dataset's annotation_3d pose_3d num is not equal to the num in multiview images. Would you please tell me wheather it is a bug or not? Like below:  
