STMT: A Spatial-Temporal Mesh Transformer for MoCap-Based Action Recognition
This addresses the problem of action recognition for researchers and practitioners by offering a more direct approach compared to skeleton-based methods, though it appears incremental as it builds on transformer architectures.
The paper tackles human action recognition from motion capture sequences by proposing STMT, a Spatial-Temporal Mesh Transformer that directly models mesh sequences, achieving state-of-the-art performance on common benchmarks.
We study the problem of human action recognition using motion capture (MoCap) sequences. Unlike existing techniques that take multiple manual steps to derive standardized skeleton representations as model input, we propose a novel Spatial-Temporal Mesh Transformer (STMT) to directly model the mesh sequences. The model uses a hierarchical transformer with intra-frame off-set attention and inter-frame self-attention. The attention mechanism allows the model to freely attend between any two vertex patches to learn non-local relationships in the spatial-temporal domain. Masked vertex modeling and future frame prediction are used as two self-supervised tasks to fully activate the bi-directional and auto-regressive attention in our hierarchical transformer. The proposed method achieves state-of-the-art performance compared to skeleton-based and point-cloud-based models on common MoCap benchmarks. Code is available at https://github.com/zgzxy001/STMT.