3DMPPE_POSENET_RELEASE
3DMPPE_POSENET_RELEASE copied to clipboard
mks0601
Official PyTorch implementation of "Camera Distance-aware Top-down Approach for 3D Multi-person Pose Estimation from a Single RGB Image", ICCV 2019
How can I get the probability of pose_3d? I find vis_kps[2,:] that always equals one in the demo.py.
First, thanks for your sharing. My work requires the use of the camera extrinsic parameters of MuPoTS, the dataset is not directly available for download and I cannot contact the...
Hi! Thanks for your great work! I have a question about the 3d key point output. The x and y of a key point is in pixels and z is...
Hello, First of all, thank you for publishing your code. I am having some trouble understanding how absolute coordinates are evaluated in MupoTS. 1- What is the threshold value used...
@wangzheallen @mks0601 hi thanks for your great work , i have few queries in ur visualization and interpretation 1. I am getting expected output for the surveillance video as shown...
Thank you very much for your contribution. I have a question. When I loaded the MuCO data, I found that the coordinates of joint_img exceeded the size of the image....
will the depth predicted from the model be influenced greatly, when imgs are token from the same camera using different pitch, row?
https://github.com/mks0601/3DMPPE_POSENET_RELEASE/blob/3f92ebaef214a0eb1574b7265e836456fbf3508a/demo/demo.py#L106 Hi mks, why it is divided by cfg.depth_dim, I don't quite understand this (pose_3d[:,2] / cfg.depth_dim * 2 - 1) * (cfg.bbox_3d_shape[0]/2). would you explain it? thx.
Amazing repo and very interesting work. I have some questions about the data formatting to introduce 3D data. This may sound very simple so I apologize in advance. When building...
It is a very excellent work! Thanks for your detailed sharing. I have learned your codes of ROOTNET and POSENET and I want to make some improvements based on your...
