| SAM 3: Segment Anything with Concepts | ✓ Link | 78.4 | | | | SAM 3 (zero-shot) (extra data) | 2025-11-20 |
| SAM 2: Segment Anything in Images and Videos | ✓ Link | 77.9 | | | | SAM2 | 2024-08-01 |
| SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction | ✓ Link | 75.3 | | | | SeC-4B (extra data) | 2025-07-21 |
| SAM2Long: Enhancing SAM 2 for Long Video Segmentation with a Training-Free Memory Tree | ✓ Link | 75.2 | | | | SAM2.1Long-L (extra data) | 2024-10-21 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 71.7 | 67.6 | 75.8 | 17.9 | Cutie+ (base, MEGA) | 2023-10-19 |
| Efficient Track Anything | ✓ Link | 71.4 | | | | EfficientTAM-S (extra data) | 2024-11-28 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 70.3 | 66.0 | 74.5 | 20.6 | Cutie+ (small, MEGA) | 2023-10-19 |
| EdgeTAM: On-Device Track Anything Model | ✓ Link | 70.0 | | | | EdgeTAM (extra data) | 2025-01-13 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 69.9 | 65.8 | 74.1 | 36.4 | Cutie (base, MEGA) | 2023-10-19 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 68.6 | 64.3 | 72.9 | 45.5 | Cutie (small, MEGA) | 2023-10-19 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 68.3 | 64.2 | 72.3 | 36.4 | Cutie (base, with mose) | 2023-10-19 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 67.4 | 63.1 | 71.7 | 45.5 | Cutie (small, with mose) | 2023-10-19 |
| Tracking Anything with Decoupled Video Segmentation | ✓ Link | 66.5 | 62.3 | 70.8 | 25.3 | DEVA (with OVIS) | 2023-09-07 |
| MobileSAM2: Lightweight Segment Anything in Images and Videos via Hypergraphical Knowledge Distillation | | 65.8 | | | | MobileSAM2-23M | 2026-09-08 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 64.0 | 60.0 | 67.9 | 36.4 | Cutie (base) | 2023-10-19 |
| Putting the Object Back into Video Object Segmentation | ✓ Link | 62.2 | 58.2 | 66.2 | 45.5 | Cutie (small) | 2023-10-19 |
| Tracking Anything with Decoupled Video Segmentation | ✓ Link | 60.0 | 55.8 | 64.3 | 25.3 | DEVA (no OVIS) | 2023-09-07 |
| Decoupling Features in Hierarchical Propagation for Video Object Segmentation | ✓ Link | 59.4 | 55.1 | 63.8 | | DeAOT | 2022-10-18 |
| XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model | ✓ Link | 57.6 | 53.3 | 62.0 | | XMem | 2022-07-14 |
| Associating Objects with Transformers for Video Object Segmentation | ✓ Link | 57.2 | 53.1 | 61.3 | | AOT | 2021-06-04 |
| SWEM: Towards Real-Time Video Object Segmentation with Sequential Weighted Expectation-Maximization | ✓ Link | 50.9 | 46.8 | 54.9 | | SWEM | 2022-08-22 |
| Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation | ✓ Link | 50.8 | 46.6 | 55.0 | | STCN | 2021-06-09 |
| Recurrent Dynamic Embedding for Video Object Segmentation | ✓ Link | 48.8 | 44.6 | 52.9 | | RDE | 2022-05-08 |
| SegGPT: Segmenting Everything In Context | ✓ Link | 45.1 | | | | SegGPT (extra data) | 2023-04-06 |
| Segment Anything Meets Point Tracking | ✓ Link | 42.9 | | | | SAM-PT (TAPIR) (extra data) | 2023-07-03 |