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 | mAP@0.1:0.7 | 60.3 | — | Unverified |
| 2 | LACP | mAP@0.1:0.7 | 52.8 | — | Unverified |
| 3 | D2-Net | mAP@0.1:0.5 | 51.4 | — | Unverified |
| 4 | SAL | mAP@0.1:0.7 | 50.6 | — | Unverified |
| 5 | PivoTAL | mAP@0.1:0.7 | 49.6 | — | Unverified |
| 6 | CASE + Zhou et al. | mAP@0.1:0.7 | 49.2 | — | Unverified |
| 7 | AFPS | mAP@0.1:0.7 | 48.5 | — | Unverified |
| 8 | P-MIL | mAP@0.1:0.7 | 47 | — | Unverified |
| 9 | AICL | mAP@0.1:0.7 | 46.4 | — | Unverified |
| 10 | TSN | mAP@0.1:0.7 | 46.2 | — | Unverified |