Dear authors, thank you for sharing such a great work with public. While trying to run your code on a sample video I encountered a problem while calculating affpts and affb values required as input to your network (you calculated and provided them for the dataset you worked on). Although I found out that affpts can be calculated using confidence heatmaps and affb can be calculated using PAFs it's not clear where you get PAFs as outputs. I checked the pose estimator you used for getting keypoints but it doesn't generate PAFs for me to calculate affb values. Would you please provide help on this issue?
Dear authors, thank you for sharing such a great work with public. While trying to run your code on a sample video I encountered a problem while calculating affpts and affb values required as input to your network (you calculated and provided them for the dataset you worked on). Although I found out that affpts can be calculated using confidence heatmaps and affb can be calculated using PAFs it's not clear where you get PAFs as outputs. I checked the pose estimator you used for getting keypoints but it doesn't generate PAFs for me to calculate affb values. Would you please provide help on this issue?