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 | MAAN | mAP@0.5 | 33.7 | — | Unverified |
| 2 | TSM | mAP@0.5 | 30.3 | — | Unverified |
| 3 | STPN | mAP@0.5 | 29.3 | — | Unverified |
| 4 | SAL | mAP@0.5:0.95 | 28.8 | — | Unverified |
| 5 | PivoTAL | mAP@0.5:0.95 | 28.1 | — | Unverified |
| 6 | AICL | mAP@0.5:0.95 | 27.6 | — | Unverified |
| 7 | AFPS | mAP@0.5:0.95 | 27.3 | — | Unverified |
| 8 | CASE | mAP@0.5:0.95 | 26.8 | — | Unverified |
| 9 | TSN | mAP@0.5:0.95 | 26.3 | — | Unverified |
| 10 | ProCL | mAP@0.5:0.95 | 26.1 | — | Unverified |