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 | BaS-Net | mAP@0.5 | 38.5 | — | Unverified |
| 2 | CleanNet | mAP@0.5 | 37.1 | — | Unverified |
| 3 | W-TALC | mAP@0.5 | 37 | — | Unverified |
| 4 | CMCS | mAP@0.5 | 36.8 | — | Unverified |
| 5 | SAL | Mean mAP | 30.8 | — | Unverified |
| 6 | AFPS | Mean mAP | 29.9 | — | Unverified |
| 7 | AICL | Mean mAP | 29.9 | — | Unverified |
| 8 | TSM | mAP@0.5 | 28.3 | — | Unverified |
| 9 | CASE | Mean mAP | 27.9 | — | Unverified |
| 10 | AutoLoc | mAP@0.5 | 27.3 | — | Unverified |