Hi! Have anybody seen PoseFormers such as PoseFormerV2 from [2303.17472] PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimation on hub? Or is there any Space for inference of that type of models?
It can be either Image-to-Video or Video-to-Video models. But these categories are not presented in Computer Vision section on hub.
Hello! PoseFormers like PoseFormerV2 are really interesting tools for 3D human pose estimation on the hub. I think these are good models. If they are not presented on the hub in the computer vision section, I think it’s worth contacting the hub administrators or moderators with a request to add such categories.