| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Referring Video Object Segmentation | JHMDB-Sentences (test) | Overall IoU0.783 | 83 | |
| Referring Video Object Segmentation | JHMDB-Sentences | Overall IoU74.4 | 56 | |
| Referring Video Segmentation | JHMDB Sentences (test) | mAP (0.5:0.95)45.8 | 35 | |
| Referring Video Object Segmentation | JHMDB Sentences | mAP46.6 | 29 | |
| Action Detection | JHMDB-21 | video-mAP@0.585.3 | 21 | |
| Pose Propagation | JHMDB | PCK@0.163.1 | 20 | |
| Human Pose Estimation | JHMDB (val) | PCK@0.159.2 | 19 | |
| Action Recognition | JHMDB Mean over 3 splits | Accuracy77.2 | 18 | |
| Video Action Detection | JHMDB21 1.0 (test) | f-mAP@0.590.2 | 17 | |
| Video label propagation | JHMDB (val) | PCK@0.163.4 | 17 | |
| Referring Video Segmentation | JHMDB Sentences | Precision @ 0.587.4 | 16 | |
| Human Pose Tracking | JHMDB (val) | PCK@.168.7 | 15 | |
| Human Pose Estimation | JHMDB | PCK@0.149.4 | 12 | |
| Action Detection | JHMDB (trimmed) | Video-mAP@0.580.1 | 12 | |
| Action Detection | JHMDB (split-1) | Brush Hair AP79.1 | 12 | |
| Action Recognition | JHMDB | Mean Per-Class Accuracy88.36 | 11 | |
| Human Pose Tracking | JHMDB (split1) | PCK @ 0.168.7 | 11 | |
| Action Detection | JHMDB (test) | F@0.561.3 | 11 | |
| Pose Keypoint Propagation | JHMDB split 1 (val) | PCK@0.168.7 | 10 | |
| Spatial Tasks | JHMDB | PCK@0.151.4 | 9 | |
| Actor and Action Segmentation | JHMDB-S (val) | oIoU72.9 | 9 | |
| Referring Video Object Segmentation | JHMDB (val) | mIoU43.7 | 9 | |
| Early Action Recognition | JHMDB | Accuracy83.5 | 9 | |
| Human Pose Estimation | Sub-JHMDB (test) | Head Accuracy99.9 | 8 | |
| Action Detection | JHMDB closed-set | F@0.592.1 | 7 |