Thanks to the authors for the great work and sharing the same. This question is open to authors or the rest. The authors mentioned they used densepose in the paper, and it is therefore understood we get the 24 part labelled output of densepose in the image under "dense" folder. However I am unable to easily generate these from the detectron2/densepose repo. I get some UV co-ordinates etc, and read somewhere that the I would correspond to the labels, but this is just not working for me. Could somebody who is able to generate the densepose outputs for this project please give me some pointers on how to go about it?
Dear @xiezhy6 /All,
Thanks to the authors for the great work and sharing the same. This question is open to authors or the rest. The authors mentioned they used densepose in the paper, and it is therefore understood we get the 24 part labelled output of densepose in the image under "dense" folder. However I am unable to easily generate these from the detectron2/densepose repo. I get some UV co-ordinates etc, and read somewhere that the I would correspond to the labels, but this is just not working for me. Could somebody who is able to generate the densepose outputs for this project please give me some pointers on how to go about it?
Thanks,