Referring Video Segmentation on Refer-Youtube-VOS (val)
63.9J IndexSgMg
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SgMgYear=2023, Processing=offline2023.12 | 63.9 | 67.4 | 65.7 | |
| OnlineReferBackbone=Swin-L, Online/Offline=online2023.07 | 61.6 | 65.5 | 63.5 | |
| OnlineReferYear=2023, Processing=online2023.12 | 61.6 | 65.5 | 63.5 | |
| ReferFormerBackbone=Video-Swin-B, Online/Offline=offline, Window Size (w)=52023.07 | 61.3 | 64.6 | 62.9 | |
| ReferFormerYear=2022, Processing=offline2023.12 | 61.3 | 64.6 | 62.9 | |
| OnlineReferBackbone=Video-Swin-B, Online/Offline=semi-online, Window Size (w)=22023.07 | 61 | 64.7 | 62.9 | |
| OnlineReferYear=2023, Processing=semi-online2023.12 | 61 | 64.7 | 62.9 | |
| ReferFormerBackbone=Swin-L, Online/Offline=offline2023.07 | 60.8 | 64 | 62.4 | |
| CITDBackbone=Ensemble, Online/Offline=offline2023.07 | 60 | 62.7 | 61.4 | |
| CITDYear=2021, Processing=offline2023.12 | 60 | 62.7 | 61.4 | |
| R2-VOSYear=2023, Processing=offline2023.12 | 59.6 | 63.1 | 61.3 | |
| R2-VOSYear=2023, Processing=online2023.12 | 58.9 | 61.5 | 60.2 | |
| TrackGPT-13BYear=2023, Processing=online2023.12 | 58.1 | 60.8 | 59.5 | |
| ReferFormerBackbone=Video-Swin-T, Online/Offline=offline, Window Size (w)=52023.07 | 58 | 60.9 | 59.4 | |
| OnlineReferBackbone=ResNet-50, Online/Offline=online2023.07 | 55.6 | 58.9 | 57.3 | |
| TrackGPT-7BYear=2023, Processing=online2023.12 | 55.3 | 57.4 | 56.4 | |
| ReferFormerBackbone=ResNet-50, Online/Offline=offline2023.07 | 54.8 | 56.5 | 55.6 | |
| MTTR2021.11 | 54 | 56.64 | 55.32 | |
| MTTRBackbone=Video-Swin-T, Online/Offline=offline, Window Size (w)=122023.07 | 54 | 56.6 | 55.3 | |
| MTTRYear=2022, Processing=offline2023.12 | 54 | 56.6 | 55.3 | |
| Ding et al.Ensemble=true2021.11 | 53.7 | 56 | 54.8 | |
| PMINet + CFBIBackbone=Ensemble, Online/Offline=offline2023.07 | 53 | 55.5 | 54.2 | |
| PMINet + CFBIYear=2021, Processing=offline2023.12 | 53 | 55.5 | 54.2 | |
| LISA-13B*Year=2023, Processing=online2023.12 | 52.1 | 53 | 52.6 | |
| LISA-7B*Year=2023, Processing=online2023.12 | 49.7 | 50.6 | 50.2 | |
| Deeply Interleaved Two-Stream Encoder2022.03 | 48.44 | 50.67 | 49.56 | |
| MLSABackbone=ResNet-50, Online/Offline=offline2023.07 | 48.4 | 50.9 | 49.7 | |
| MLSAYear=2022, Processing=offline2023.12 | 48.4 | 50.9 | 49.7 | |
| LBDTYear=2022, Processing=online2023.12 | 48.2 | 50.6 | 49.4 | |
| LBDT-4Number of LBDT layers=42022.06 | 48.18 | 50.57 | 49.38 | |
| YOFOYear=2022, Processing=online2023.12 | 47.5 | 49.7 | 48.6 | |
| CMPC-V2022.03 | 45.64 | 49.32 | 47.48 | |
| CMPC-VBackbone=I3D2021.11 | 45.64 | 49.32 | 47.48 | |
| CMPC-VPub.=TPAMI212022.06 | 45.64 | 49.32 | 47.48 | |
| CMPC-VYear=2021, Processing=offline2023.12 | 45.6 | 49.3 | 47.5 | |
| URVOS2022.03 | 45.27 | 49.19 | 47.23 | |
| URVOS2021.11 | 45.27 | 49.19 | 47.23 | |
| URVOSPub.=ECCV20, Inference iterations=multiple iterations2022.06 | 45.27 | 49.19 | 47.23 | |
| URVOSBackbone=ResNet-50, Online/Offline=online2023.07 | 45.2 | 49.1 | 47.2 | |
| URVOSYear=2020, Processing=online2023.12 | 45.2 | 49.1 | 47.2 | |
| URVOSPub.=ECCV20, Inference iterations=single stage2022.06 | 41.34 | — | — | |
| CMSA + RNNBackbone=ResNet-50, Online/Offline=online2023.07 | 34.7 | 38 | 36.4 | |
| CMSABackbone=ResNet-50, Online/Offline=online2023.07 | 33.3 | 36.5 | 34.9 | |
| CMSAYear=2019, Processing=online2023.12 | 33.3 | 36.5 | 34.9 |