fenaux

And that would avoid compatibility issue with mmpose 1.x https://github.com/ViTAE-Transformer/ViTPose/issues/108

I tested the following process on colab and it works with the two pictures of the main page plus a video first clone the repo git_repo_url = 'https://github.com/openpifpaf/openpifpaf.git' !git clone...

May be you can find the answer in this paper https://arxiv.org/pdf/2111.06377.pdf Just keep the encoder as a backbone and create a head for object detection. Case of R-CNN is described...