This is an official implementation of our CVPR 2021 paper "Deep Dual Consecutive Network for Human Pose Estimation" (https://openaccess.thecvf.com/content/CVPR2021/papers/Liu_Deep_Dual_Consecutive_Network_for_Human_Pose_Estimation_CVPR_2021_paper.pdf)
Why not let the model directly output 15 joints corresponding to Posetrack dataset, but output 17 joints corresponding to COCO dataset, and then convert to posetrack?
Why not let the model directly output 15 joints corresponding to Posetrack dataset, but output 17 joints corresponding to COCO dataset, and then convert to posetrack?