Closed lxtGH closed 4 years ago
Just use pre-trained model on COCO, and set loss zero for not annotated l_ear and r_ear. When fine-tuning, slight different joint (head_bottom and head_top) can be handled by learning.
Hi! Thanks for your kind reply!!!!. When doing the evaluation how to handle the not anotated l_ear and r_ear.@mks0601
I cannot understand your question. Could you make it clear?
@mks0601 Hi! Sorry for that. I update my moments.
I think so, but cannot remember clearly. Sorry :(
Hi! Thanks for your repo. The number of joints in posetrack is a little different from COCO. How did you handle the missing joints like left_ear and right_ear. Also the different joint head_bottom and head_top?