Daniil-Osokin / gccpm-look-into-person-cvpr19.pytorch

Fast and accurate single-person pose estimation, ranked 10th at CVPR'19 LIP challenge. Contains implementation of "Global Context for Convolutional Pose Machines" paper.
Apache License 2.0
168 stars 28 forks source link