Mimi Liao
Mimi Liao
Maybe it is because the Human3.6M SMPL ground truth provided by Mosh is no longer available. Reference: https://github.com/nkolot/ProHMR/tree/master/dataset_preprocessing > Human3.6M: Unfortunately, due to license limitations, we are not allowed to...
In https://github.com/hongsukchoi/Pose2Mesh_RELEASE/blob/7f24836c36dfdd52be6735505f44af11ec97e666/data/PW3D/dataset.py#L93, the ground truth 3DPW vertices are generated with camera trans.
@hongsukchoi Sorry, I found that the predicted camera parameter I mentioned is in your another repo: https://github.com/hongsukchoi/3DCrowdNet_RELEASE/blob/6e773064c8d6950b382f66d76b615aada4f2594b/main/model.py#L65 Also, what do you mean in this sentence: "Other methods assuming a single...
Hi, In my understanding, there are world coordinates (3D), camera coordinates (3D), and pixel coordinates (2D). And the model's output is already in camera coordinates, so the predicted camera is...