YouTube-VIS
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Video Instance Segmentation | YouTube-VIS 2019 (val) | AP69.4 | 620 | |
| Video Instance Segmentation | YouTube-VIS 2021 (val) | AP65.3 | 372 | |
| Video Instance Segmentation | YouTube-VIS 2019 | AP77.9 | 123 | |
| Video Instance Segmentation | YouTube-VIS (val) | AP47.6 | 118 | |
| Video Instance Segmentation | YouTube-VIS 2021 | AP72.3 | 99 | |
| Video Instance Segmentation | YouTube-VIS 2022 (val) | AP52.4 | 37 | |
| Video Instance Segmentation | YouTube-VIS long videos 2022 (val) | AP45.9 | 15 | |
| Video Instance Segmentation | YouTube-VIS 2019 (test) | AP61.6 | 13 | |
| Video Instance Segmentation | YouTube VIS 2021 (test) | mAP54.1 | 13 | |
| Video Instance Segmentation | YouTube-VIS (train val) | AP5078.24 | 11 | |
| Video Object-Centric Learning | YouTube-VIS | FG-ARI44.8 | 9 | |
| object dynamics prediction | YouTube-VIS 2021 (test) | FG-ARI42.9 | 9 | |
| Video Instance Segmentation | YouTube-VIS 2019 (Challenge) | mAP46.7 | 9 | |
| Few-Shot Video Instance Segmentation | YouTube-VIS | mIoU (Fold 1)48.4 | 7 | |
| Object dynamics prediction | YouTube-VIS | FG-ARI66.6 | 7 | |
| Video Semantic Segmentation | YouTube-VIS 2021 | mAP44.2 | 7 | |
| Video Instance Segmentation | YouTube-VIS (test) | AP0.323 | 7 | |
| Video Object Discovery and Tracking | YouTube-VIS HQ | ARIfg76.6 | 5 | |
| Video Object-Centric Learning | YouTube-VIS (test) | FG-ARI44.8 | 5 | |
| Object tracking | YouTube-VIS first 24 frames | ARI23 | 4 | |
| Video Instance Segmentation | YouTube-VIS 2022 | AP5035.3 | 2 |
Showing 21 of 21 rows