USDRL: Unified Skeleton-Based Dense Representation Learning with Multi-Grained Feature Decorrelation
About
Contrastive learning has achieved great success in skeleton-based representation learning recently. However, the prevailing methods are predominantly negative-based, necessitating additional momentum encoder and memory bank to get negative samples, which increases the difficulty of model training. Furthermore, these methods primarily concentrate on learning a global representation for recognition and retrieval tasks, while overlooking the rich and detailed local representations that are crucial for dense prediction tasks. To alleviate these issues, we introduce a Unified Skeleton-based Dense Representation Learning framework based on feature decorrelation, called USDRL, which employs feature decorrelation across temporal, spatial, and instance domains in a multi-grained manner to reduce redundancy among dimensions of the representations to maximize information extraction from features. Additionally, we design a Dense Spatio-Temporal Encoder (DSTE) to capture fine-grained action representations effectively, thereby enhancing the performance of dense prediction tasks. Comprehensive experiments, conducted on the benchmarks NTU-60, NTU-120, PKU-MMD I, and PKU-MMD II, across diverse downstream tasks including action recognition, action retrieval, and action detection, conclusively demonstrate that our approach significantly outperforms the current state-of-the-art (SOTA) approaches. Our code and models are available at https://github.com/wengwanjiang/USDRL.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Action Recognition | NTU RGB+D 120 (X-set) | Accuracy80.6 | 661 | |
| Skeleton-based Action Recognition | NTU 60 (X-sub) | Accuracy87.1 | 220 | |
| Skeleton-based Action Recognition | NTU RGB+D 120 (X-set) | Top-1 Accuracy80.6 | 184 | |
| Action Recognition | NTU RGB+D X-View 60 | Accuracy93.2 | 172 | |
| Skeleton-based Action Recognition | NTU 120 (X-sub) | -- | 139 | |
| Skeleton-based Action Recognition | NTU RGB+D 60 (X-View) | Top-1 Accuracy93.2 | 126 | |
| Action Recognition | NTU-120 (cross-subject (xsub)) | Accuracy79.3 | 82 | |
| Action Recognition | PKU-MMD II (xsub) | Accuracy59.7 | 42 | |
| Action Recognition | NTU-60 (xsub) | Accuracy87.1 | 40 | |
| Skeleton-based Action Recognition | PKU-MMD II (x-sub) | Top-1 Acc59.7 | 21 |