T-MOR: Learning Motion-Aware Skeleton Representations for Human Action Recognition
About
Vision-language models such as CLIP have recently achieved strong performance on a wide range of visual understanding tasks. However, most existing models rely primarily on appearance-level supervision from images or videos, and do not explicitly model human motion, which is essential for fine-grained and human-centric action recognition task as actions are defined by temporally structured and physically grounded body movements. To address this problem, we propose Transferable skeleton MOtion Representation (T-MOR), a motion-aware framework that learns transferable action representations from skeleton sequences with the aid of video and language supervision during training. T-MOR adopts a multi-modal contrastive learning scheme that aligns skeleton motion with visual and textual representations, while performing inference using only lightweight skeleton inputs. To support large-scale pre-training, we construct PoseCap-1M, a new dataset that contains over one million synchronized video, skeleton, and text triplets covering diverse human activities. We evaluate T-MOR on a range of human-centric action recognition benchmarks, including action classification and frame-wise temporal detection. Experimental results show that T-MOR consistently improves performance across multiple datasets, such as Toyota Smarthome, Penn Action, UAV-Human, TSU, and Charades. In addition, T-MOR demonstrates strong generalization ability in few-shot and zero-shot settings, highlighting the effectiveness of motion-centric and embodied representations for transferable action understanding.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Action Recognition | UAV-Human (CSv2) | Accuracy70.8 | 30 | |
| Action Classification | Smarthome (cross-view CV2) | Accuracy66.7 | 28 | |
| Action Classification | Smarthome CS | Top-1 Accuracy66.2 | 21 | |
| Action Classification | Penn-Action | Top-1 Accuracy98.2 | 19 | |
| Action Detection | Charades | mAP (%)26 | 18 | |
| Action Detection | Toyota Smarthome Untrimmed (TSU) (Cross-Subject (CS)) | CS Accuracy38.3 | 11 | |
| Action Classification | UAV-Human CS1 | CS1 Accuracy44.4 | 10 | |
| Action Detection | PKU-MMD (CS) | mAP@IoU 0.194.3 | 8 |