Chirality Nets for Human Pose Regression
arXiv:1911.00029
Abstract
We propose Chirality Nets, a family of deep nets that is equivariant to the "chirality transform," i.e., the transformation to create a chiral pair. Through parameter sharing, odd and even symmetry, we propose and prove variants of standard building blocks of deep nets that satisfy the equivariance property, including fully connected layers, convolutional layers, batch-normalization, and LSTM/GRU cells. The proposed layers lead to a more data efficient representation and a reduction in computation by exploiting symmetry. We evaluate chirality nets on the task of human pose regression, which naturally exploits the left/right mirroring of the human body. We study three pose regression tasks: 3D pose estimation from video, 2D pose forecasting, and skeleton based activity recognition. Our approach achieves/matches state-of-the-art results, with more significant gains on small datasets and limited-data settings.
Accepted to NeurIPS2019
Cited by in corpus (5)
- Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations
- Anatomy-aware 3D Human Pose Estimation with Bone-based Pose Decomposition
- On discrete symmetries of robotics systems: A group-theoretic and data-driven analysis
- Pose Refinement Graph Convolutional Network for Skeleton-based Action Recognition
- Occlusion-Invariant Rotation-Equivariant Semi-Supervised Depth Based Cross-View Gait Pose Estimation