Weakly-supervised 3D Pose Transfer with Keypoints

We present a new method capable of transferring pose for 3D articulated objects in a weakly-supervised way.

The main challenges of 3D pose transfer are: 1) Lack of paired training data with different characters perform- ing the same pose; 2) Disentangling pose and shape infor- mation from the target mesh; 3) Difficulty in applying to meshes with different topologies. We thus propose a novel weakly-supervised keypoint-based framework to overcome these difficulties. Specifically, we use a topology-agnostic keypoint detector with inverse kinematics to compute trans- formations between the source and target meshes. Our method only requires supervision on the keypoints, can be applied to meshes with different topologies and is shape- invariant for the target which allows extraction of pose-only information from the target meshes without transferring shape information. We further design a cycle reconstruction to perform self-supervised pose transfer without the need for ground truth deformed mesh with the same pose and shape as the target and source, respectively. We evaluate our approach on benchmark human and animal datasets, where we achieve superior performance compared to the state-of-the-art unsupervised approaches and even compa- rable performance with the fully supervised approaches. We test on the more challenging Mixamo dataset to verify our approach’s ability in handling meshes with different topolo- gies and complex clothes. Cross-dataset evaluation further shows the strong generalization ability of our approach.