Referring Video Object Segmentation on Ref-YouTube-VOS 1.0 (val)
66.5J&FVD-IT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| VD-ITBackbone=Video Diffusion, Pre-training=RefCOCO/+/g2024.03 | 66.5 | 64.4 | 68.5 | 21 | |
| SgMgBackbone=Video-Swin-B, Pre-training=RefCOCO/+/g2024.03 | 65.7 | 63.9 | 67.4 | 40 | |
| VD-ITBackbone=Video Diffusion2024.03 | 64.8 | 63.1 | 66.6 | 21 | |
| ReferFormerBackbone=Video-Swin-B, Pre-training=RefCOCO/+/g2024.03 | 62.9 | 61.3 | 64.6 | 33 | |
| OnlineReferBackbone=Video-Swin-B, Pre-training=RefCOCO/+/g2024.03 | 62.9 | 61 | 64.7 | — | |
| SgMgBackbone=Video-Swin-T, Pre-training=RefCOCO/+/g2024.03 | 62 | 60.4 | 63.5 | 65 | |
| SgMgBackbone=Video-Swin-B, Note=Evaluated using official code2024.03 | 61.6 | 59.7 | 63.5 | 40 | |
| ReferFormerBackbone=Video-Swin-T, Pre-training=RefCOCO/+/g2024.03 | 59.4 | 58 | 60.9 | 50 | |
| SgMgBackbone=Video-Swin-T2024.03 | 58.9 | 57.7 | 60 | 65 | |
| ReferFormerBackbone=Video-Swin-T2024.03 | 56 | 54.8 | 57.3 | 50 | |
| MANetBackbone=Video-Swin-T2024.03 | 55.6 | 54.8 | 56.5 | — | |
| MTTRBackbone=Video-Swin-T2024.03 | 55.3 | 54 | 56.6 | — | |
| PMINetBackbone=ResNeSt-1012024.03 | 53 | 51.5 | 54.5 | — | |
| MLRLBackbone=ResNet-502024.03 | 49.7 | 48.4 | 51 | — | |
| LBDTBackbone=ResNet-502024.03 | 49.4 | 48.2 | 50.6 | — | |
| YOFOBackbone=ResNet-502024.03 | 48.6 | 47.5 | 49.7 | 10 | |
| CMPC-VBackbone=I3D2024.03 | 47.5 | 45.6 | 49.3 | — | |
| URVOSBackbone=ResNet-502024.03 | 47.2 | 45.3 | 49.2 | — | |
| CMSABackbone=ResNet-502024.03 | 36.4 | 34.8 | 38.1 | — |