Referring Video Object Segmentation on A2D-Sentences
82.1oIoUReferDINO
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| ReferDINOBackbone=Swin-B2025.01 | 82.1 | 61.1 | 73.6 | — | — | — | — | — | |
| LMPM++Backbone=Video-Swin-Base2025.12 | 81.8 | 60.7 | 73.9 | — | — | — | — | — | |
| VD-ITBackbone=Video Diffusion2024.03 | 81.5 | 61.4 | 73.2 | — | — | — | — | — | |
| LoShBackbone=Swin-B2025.01 | 81.2 | 59.9 | 73.1 | — | — | — | — | — | |
| LoShBackbone=Video-Swin-Base2025.12 | 81.2 | 59.9 | 73.1 | — | — | — | — | — | |
| DsHmpBackbone=Swin-B2025.01 | 81.1 | 59.8 | 72.9 | — | — | — | — | — | |
| DsHmpBackbone=Video-Swin-Base2025.12 | 81.1 | 59.8 | 72.9 | — | — | — | — | — | |
| G-DINO-SHBackbone=Swin-B2025.01 | 80.9 | 58.4 | 72.6 | — | — | — | — | — | |
| SOCBackbone=Swin-B2025.01 | 80.7 | 57.3 | 72.5 | — | — | — | — | — | |
| SOCBackbone=Video-Swin-Base2025.12 | 80.7 | 57.3 | 72.5 | — | — | — | — | — | |
| ReferDINOBackbone=Swin-T2025.01 | 80.2 | 58.9 | 72.3 | — | — | — | — | — | |
| G-DINO-DHBackbone=Swin-B2025.01 | 80.2 | 58.7 | 71.8 | — | — | — | — | — | |
| HTRBackbone=VideoSwin-B2024.03 | 80.1 | 59 | 72.6 | 85.2 | 83.2 | 77.9 | 62.4 | 25.5 | |
| SgMgBackbone=Swin-B2025.01 | 79.9 | 58.5 | 72 | — | — | — | — | — | |
| SgMgBackbone=Video-Swin-B2024.03 | 79.9 | 58.5 | 72 | — | — | — | — | — | |
| SgMgBackbone=Video-Swin-Base2025.12 | 79.9 | 58.5 | 72 | — | — | — | — | — | |
| OnlineReferBackbone=VideoSwin-B2024.03 | 79.6 | — | 70.5 | 83.1 | 80.2 | 73.4 | 56.8 | 21.7 | |
| OnlineReferBackbone=Video-Swin-Base2025.12 | 79.6 | — | 70.5 | — | — | — | — | — | |
| HTMLBackbone=Swin-B2025.01 | 79.5 | 56.7 | 71.2 | — | — | — | — | — | |
| HTMLBackbone=VideoSwin-B2024.03 | 79.5 | 56.7 | 71.2 | 84 | 81.5 | 75.8 | 59.2 | 22.8 | |
| HTMLBackbone=Video-Swin-Base2025.12 | 79.5 | 56.7 | 71.2 | — | — | — | — | — | |
| LMPM++Backbone=Video-Swin-Tiny2025.12 | 79.4 | 57.8 | 71.6 | — | — | — | — | — | |
| LoShBackbone=Swin-T2025.01 | 79.3 | 57.6 | 71.6 | — | — | — | — | — | |
| LoShBackbone=Video-Swin-Tiny2025.12 | 79.3 | 57.6 | 71.6 | — | — | — | — | — | |
| DsHmpBackbone=Swin-T2025.01 | 79 | 57.2 | 71.3 | — | — | — | — | — | |
| G-DINO-SHBackbone=Swin-T2025.01 | 79 | 57.8 | 71 | — | — | — | — | — | |
| DsHmpBackbone=Video-Swin-Tiny2025.12 | 79 | 57.2 | 71.3 | — | — | — | — | — | |
| VLTBackbone=VideoSwin-B2024.03 | 78.8 | 55.6 | 70.5 | 83.1 | 80.7 | 74.5 | 58.5 | 21.7 | |
| ReferFormerBackbone=Swin-B2025.01 | 78.6 | 55 | 70.3 | — | — | — | — | — | |
| ReferFormerBackbone=Video-Swin-B2024.03 | 78.6 | 55 | 70.3 | — | — | — | — | — | |
| ReferFormerBackbone=VideoSwin-B2024.03 | 78.6 | 55 | 70.3 | 83.1 | 80.4 | 74.1 | 57.9 | 21.2 | |
| ReferFormerBackbone=Video-Swin-Base2025.12 | 78.6 | 55 | 70.3 | — | — | — | — | — | |
| SOCBackbone=Swin-T2025.01 | 78.3 | 54.8 | 70.6 | — | — | — | — | — | |
| SOCBackbone=Video-Swin-Tiny2025.12 | 78.3 | 54.8 | 70.6 | — | — | — | — | — | |
| G-DINO-DHBackbone=Swin-T2025.01 | 78.2 | 57.3 | 69.9 | — | — | — | — | — | |
| SgMgBackbone=Swin-T2025.01 | 78 | 56.1 | 70.4 | — | — | — | — | — | |
| SgMgBackbone=Video-Swin-T2024.03 | 78 | 56.1 | 70.4 | — | — | — | — | — | |
| SgMgBackbone=Video-Swin-Tiny2025.12 | 78 | 56.1 | 70.4 | — | — | — | — | — | |
| ReferFormerBackbone=Swin-T2025.01 | 77.6 | 52.8 | 69.6 | — | — | — | — | — | |
| HTMLBackbone=Swin-T2025.01 | 77.6 | 53.4 | 69.2 | — | — | — | — | — | |
| ReferFormerBackbone=Video-Swin-T2024.03 | 77.6 | 52.8 | 69.6 | — | — | — | — | — | |
| ReferFormerBackbone=Video-Swin-Tiny2025.12 | 77.6 | 52.8 | 69.6 | — | — | — | — | — | |
| HTMLBackbone=Video-Swin-Tiny2025.12 | 77.6 | 53.4 | 69.2 | — | — | — | — | — | |
| MTTRBackbone=Video-Swin-T2024.03 | 72 | 46.1 | 64 | — | — | — | — | — | |
| MTTRBackbone=VideoSwin-T2024.03 | 72 | 46.1 | 64 | 75.4 | 71.2 | 63.8 | 48.5 | 16.9 | |
| MTTRBackbone=Video-Swin-Tiny2025.12 | 72 | 46.1 | 64 | — | — | — | — | — | |
| CMPC-VBackbone=I3D2024.03 | 65.3 | 40.4 | 57.3 | — | — | — | — | — | |
| CMPC-VBackbone=I3D2024.03 | 65.3 | 40.4 | 57.3 | 65.5 | 59.2 | 50.6 | 34.2 | 9.8 | |
| ClawCraneNetBackbone=ResNet-50/1012024.03 | 63.1 | — | 59.9 | — | — | — | — | — | |
| ClawCraneNetBackbone=ResNet-50/1012024.03 | 63.1 | — | 59.9 | 70.4 | 67.7 | 61.7 | 48.9 | 17.1 | |
| CMSA + CFSABackbone=ResNet-1012024.03 | 61.8 | — | 43.2 | 48.7 | 43.1 | 35.8 | 23.1 | 5.2 | |
| ACANBackbone=I3D2024.03 | 60.1 | 27.4 | 49 | — | — | — | — | — | |
| ACANBackbone=I3D2024.03 | 60.1 | 27.4 | 49 | 55.7 | 45.9 | 31.9 | 16 | 2 | |
| Gavrilyuk et al.Backbone=I3D2024.03 | 53.6 | 19.8 | 42.1 | — | — | — | — | — | |
| Gavrilyuk et al.Backbone=I3D2024.03 | 53.6 | 19.8 | 42.1 | 47.5 | 34.7 | 21.1 | 8 | 0.2 | |
| WSRVOSModel Type=Weakly-supervised RVOS2026.04 | 48.2 | — | 44 | — | — | — | — | — | |
| Hu et al.Backbone=VGG-162024.03 | 47.4 | 13.2 | 35 | — | — | — | — | — | |
| Hu et al.Backbone=VGG-162024.03 | 47.4 | 13.2 | 35 | 34.8 | 23.6 | 13.3 | 3.3 | 0.1 | |
| Qwen3-VL-8BModel Type=Pre-trained MLLM2026.04 | 24.5 | — | 19.3 | — | — | — | — | — | |
| VideoLLaMA3-7BModel Type=Pre-trained MLLM2026.04 | 18.2 | — | 14.5 | — | — | — | — | — | |
| Video-LLaVA-7BModel Type=Pre-trained MLLM2026.04 | 15.8 | — | 11.1 | — | — | — | — | — |