Open ybk66 opened 1 year ago
To my knowledge, the code as-is does not support keypoint detection for pose estimation (see also #56 for additional ideas), but following KAPAO, we simply adapted GLIP to the keypoint task by treating them as small objects.
I repeatedly see "keypoint" in the code and I would like to ask the author and everyone here if this model supports keypoint detection similar to YOLOv8?