Weakly Supervised Action Localization
In this task, the training data consists of videos with a list of activities in them without any temporal boundary annotations. However, while testing, given a video, the algorithm should recognize the activities in the video and also provide the start and end time.
Papers
Showing 1–10 of 55 papers
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | HR-Pro | avg-mAP (0.3-0.7) | 51.1 | — | Unverified |
| 2 | LACP | mAP@0.5 | 45.3 | — | Unverified |
| 3 | LACP | avg-mAP (0.3-0.7) | 44.5 | — | Unverified |
| 4 | ProCL | mAP@0.5 | 40.5 | — | Unverified |
| 5 | P-MIL | mAP@0.5 | 40 | — | Unverified |
| 6 | P-MIL | avg-mAP (0.3-0.7) | 38 | — | Unverified |
| 7 | D2-Net | mAP@0.5 | 35.9 | — | Unverified |
| 8 | Ju et al. | mAP@0.5 | 35.9 | — | Unverified |
| 9 | Ju et al. | avg-mAP (0.3-0.7) | 35.4 | — | Unverified |
| 10 | ACM-Net | mAP@0.5 | 34.6 | — | Unverified |