Vision-Language Tracking on OTB 99
76.4AUCVLTTT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VLTTTReference Setting=NL+BBOX2024.01 | 76.4 | 93.1 | |
| VLTT2023.04 | 76.4 | 93.1 | |
| VLTTTInput Modality=Vision-language2026.01 | 76.4 | 93.1 | |
| DecoupleTNL2023.04 | 73.8 | 94.8 | |
| DecoupleTNL2024.12 | 73.8 | 94.8 | |
| DecoupleTNLReal-time vs Non-real-time=Non-real-time, GPU speed (fps)=322026.03 | 73.8 | 94.8 | |
| SUTrackModel Scale=L, Resolution=2242024.12 | 72.7 | 94.4 | |
| SDTrackSource=ESWA 20262026.06 | 72.7 | 94.2 | |
| Open_VLTrackSource=-2026.06 | 72.4 | 94.8 | |
| ADTrackSource=AAAI 20262026.06 | 72.3 | 93.9 | |
| SemaTrackSource=TVC 20262026.06 | 72.2 | 94.5 | |
| SeqTrackv2-B384Backbone=Base, Input Size=3842023.04 | 71.8 | 93.4 | |
| SeqTrackv2-L384Backbone=Large, Input Size=3842023.04 | 71.4 | 93.6 | |
| SeqTrackv2Model Scale=L, Resolution=3842024.12 | 71.4 | 93.6 | |
| GLAD-L384Input Modality=Vision-language, Input Resolution=3842026.01 | 71.4 | 93.2 | |
| SeqTrackv2-B256Backbone=Base, Input Size=2562023.04 | 71.2 | 93.9 | |
| SUTrackModel Scale=L, Resolution=3842024.12 | 71.2 | 93.1 | |
| SeqTrackv2Model Scale=B, Resolution=2562024.12 | 71.2 | 93.9 | |
| DMTrack-384Input Modality=Vision-language, Input Resolution=3842026.01 | 71.2 | 92.3 | |
| SeqTrackv2Real-time vs Non-real-time=Non-real-time, GPU speed (fps)=23, CPU speed (fps)=2, AGX speed (fps)=52026.03 | 71.2 | 93.9 | |
| ABOTrackSource=JVCIR 20262026.06 | 71.2 | 91.1 | |
| RrmTrackSource=TMM 20262026.06 | 71.2 | 93.7 | |
| UVLTrack-LReference Setting=NL+BBOX2024.01 | 71.1 | 92 | |
| CapsuleTNL2023.04 | 71.1 | 92.4 | |
| CapsuleTNL2024.12 | 71.1 | 92.4 | |
| ReasoningTrackSource=arXiv 20252026.06 | 71.1 | 95.6 | |
| All-in-oneInput Modality=Vision-language2026.01 | 71 | 93 | |
| ATSTrackSource=arXiv 20252026.06 | 71 | 94.4 | |
| DUTrackSource=CVPR 20252026.06 | 70.9 | 93.9 | |
| SUTrackModel Scale=B, Resolution=2242024.12 | 70.8 | 93.4 | |
| MMTrackInput Modality=Vision-language2026.01 | 70.5 | 91.8 | |
| MMTrack-384Source=TCSVT 20232026.06 | 70.5 | 91.8 | |
| SeqTrackv2-L256Backbone=Large, Input Size=2562023.04 | 70.3 | 91 | |
| Zhao et al.2023.04 | 69.9 | 91.2 | |
| Zhao et al.2024.12 | 69.9 | 91.2 | |
| TransVLTInput Modality=Vision-language2026.01 | 69.9 | 90.6 | |
| Zhao et al.Real-time vs Non-real-time=Non-real-time, GPU speed (fps)=362026.03 | 69.9 | 91.2 | |
| TransVLTSource=PRL 20232026.06 | 69.9 | 91.2 | |
| SUTrackModel Scale=B, Resolution=3842024.12 | 69.7 | 91.2 | |
| OneTracker2024.12 | 69.7 | 91.5 | |
| OneTrackerReal-time vs Non-real-time=Non-real-time2026.03 | 69.7 | 91.5 | |
| GLAD-B256Input Modality=Vision-language, Input Resolution=2562026.01 | 69.6 | 92.4 | |
| CiteTrackerReal-time vs Non-real-time=Non-real-time, GPU speed (fps)=22, CPU speed (fps)=2, AGX speed (fps)=42026.03 | 69.6 | 85.1 | |
| GLADSource=IJCV 20262026.06 | 69.6 | 92.4 | |
| DecoupleTNLInput Modality=Vision-language2026.01 | 69.5 | 92.8 | |
| UVLTrack-BReference Setting=NL+BBOX2024.01 | 69.3 | 89.9 | |
| DMTrack-256Input Modality=Vision-language, Input Resolution=2562026.01 | 69.3 | 90.9 | |
| UVLTrack-BReal-time vs Non-real-time=Non-real-time, GPU speed (fps)=52, CPU speed (fps)=4, AGX speed (fps)=62026.03 | 69.3 | 89.9 | |
| ATTrackSource=MM 20242026.06 | 69.3 | 90.3 | |
| UVLTrack-BSource=AAAI 20242026.06 | 69.3 | 89.9 | |
| TDCLSource=ECAI 20242026.06 | 69.1 | 90.2 | |
| Li et al.2023.04 | 69 | 91 | |
| Li et al.2024.12 | 69 | 91 | |
| Li et al.Input Modality=Vision-language2026.01 | 69 | 91 | |
| TNL2K-2Reference Setting=NL+BBOX2024.01 | 68 | 88 | |
| TNL2K-22023.04 | 68 | 88 | |
| TNL2K-22024.12 | 68 | 88 | |
| TNL2K-IIInput Modality=Vision-language2026.01 | 68 | 88 | |
| TNL2k-IISource=CVPR 20212026.06 | 68 | 88 | |
| SUTrackModel Scale=T, Resolution=2242024.12 | 67.4 | 88.6 | |
| SUTrack-TReal-time vs Non-real-time=Real-time, GPU speed (fps)=100, CPU speed (fps)=23, AGX speed (fps)=342026.03 | 67.4 | 88.6 | |
| Feng et al.2023.04 | 67 | 73 | |
| Feng et al.2024.12 | 67 | 73 | |
| QueryNLTInput Modality=Vision-language2026.01 | 66.7 | 88.2 | |
| QueryNLTSource=CVPR 20242026.06 | 66.7 | 88.2 | |
| SNLTReference Setting=NL+BBOX2024.01 | 66.6 | 80.4 | |
| SNLT2023.04 | 66.6 | 80.4 | |
| SNLT2024.12 | 66.6 | 80.4 | |
| SNLTInput Modality=Vision-language2026.01 | 66.6 | 84.8 | |
| SNLTReal-time vs Non-real-time=Non-real-time, GPU speed (fps)=502026.03 | 66.6 | 80.4 | |
| SNLTSource=CVPR 20212026.06 | 66.6 | 84.8 | |
| OSDTSource=TCSVT 20242026.06 | 66.2 | 86.7 | |
| Wang et al.2023.04 | 65.8 | 89.1 | |
| Wang et al.2024.12 | 65.8 | 89.1 | |
| JointNLTReference Setting=NL+BBOX2024.01 | 65.3 | 85.6 | |
| JointNLT2023.04 | 65.3 | 85.6 | |
| JointNLT2024.12 | 65.3 | 85.6 | |
| JointNLTInput Modality=Vision-language2026.01 | 65.3 | 85.6 | |
| JointNLTReal-time vs Non-real-time=Non-real-time, GPU speed (fps)=39, CPU speed (fps)=3, AGX speed (fps)=52026.03 | 65.3 | 85.6 | |
| JointNLTSource=CVPR 20232026.06 | 65.3 | 85.6 | |
| UETrack-TReal-time vs Non-real-time=Real-time, GPU speed (fps)=221, CPU speed (fps)=83, AGX speed (fps)=772026.03 | 64.8 | 84.7 | |
| UVLTrack-LReference Setting=NL2024.01 | 63.5 | 83.2 | |
| UETrack-SReal-time vs Non-real-time=Real-time, GPU speed (fps)=183, CPU speed (fps)=68, AGX speed (fps)=672026.03 | 63.1 | 81.9 | |
| UETrack-BReal-time vs Non-real-time=Real-time, GPU speed (fps)=163, CPU speed (fps)=56, AGX speed (fps)=602026.03 | 61.3 | 79.7 | |
| RTTNLD2023.04 | 61 | 79 | |
| RTTNLD2024.12 | 61 | 79 | |
| RTTNLDInput Modality=Vision-language2026.01 | 61 | 79 | |
| RTTNLDReal-time vs Non-real-time=Non-real-time, GPU speed (fps)=302026.03 | 61 | 79 | |
| LSTMTrackSource=WACV 20202026.06 | 61 | 79 | |
| UVLTrack-BReference Setting=NL2024.01 | 60.1 | 79.1 | |
| JointNLTReference Setting=NL2024.01 | 59.2 | 77.6 | |
| GTIReference Setting=NL2024.01 | 58.1 | 73.2 | |
| GTI2023.04 | 58.1 | 73.2 | |
| GTI2024.12 | 58.1 | 73.2 | |
| GTIInput Modality=Vision-language2026.01 | 58.1 | 73.2 | |
| GTISource=TCSVT 20212026.06 | 57.1 | 73.2 | |
| TNLS2023.04 | 55.3 | 72.3 | |
| TNLS2024.12 | 55.3 | 72.3 | |
| TNLS-IIIReference Setting=NL+BBOX2024.01 | 55 | 72 | |
| CTRNLTReference Setting=NL2024.01 | 53 | 72 |