Visual Object Tracking on VOT 2020 (test)
0.729EAOSAM2.1++
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SAM2.1++2024.11 | 0.729 | 0.799 | 0.961 | — | |
| SAM2.12024.11 | 0.681 | 0.778 | 0.941 | — | |
| SwinB-DeAOT-Lbackbone=Swin-B, variant=Large2022.10 | 0.622 | — | — | 0.559 | |
| R50-DeAOT-Lbackbone=ResNet-50, variant=Large2022.10 | 0.613 | — | — | 0.571 | |
| ODTrack-LPost-processing=Alpha-Refine2024.01 | 0.605 | 0.761 | 0.902 | — | |
| ODTrack2024.11 | 0.605 | 0.761 | 0.902 | — | |
| DeAOT-Svariant=Small2022.10 | 0.593 | — | — | 0.559 | |
| DeAOT-Lvariant=Large2022.10 | 0.591 | — | — | 0.554 | |
| SwinB-AOT-Lbackbone=Swin-B, variant=Large2022.10 | 0.586 | — | — | 0.523 | |
| MixViT-L2023.02 | 0.584 | 0.755 | 0.89 | — | |
| MixViT-L+AR2024.11 | 0.584 | 0.755 | 0.89 | — | |
| ODTrack-BPost-processing=Alpha-Refine2024.01 | 0.581 | 0.764 | 0.877 | — | |
| AOT-Lvariant=Large2022.10 | 0.574 | — | — | 0.56 | |
| DeAOT-Bvariant=Base2022.10 | 0.571 | — | — | 0.542 | |
| R50-AOT-Lbackbone=ResNet-50, variant=Large2022.10 | 0.569 | — | — | 0.54 | |
| MixViT-LPre-training=ConvMAE2023.02 | 0.567 | 0.747 | 0.87 | — | |
| SeqTrack-L2024.11 | 0.561 | — | — | — | |
| MixFormer-LPre-trained=ImageNet-22k, Mask prediction=Alpha-Refine [58]2022.03 | 0.555 | 0.762 | 0.855 | — | |
| MixFormer-Lbackbone=Large2022.10 | 0.555 | — | — | — | |
| MixCvT-L2023.02 | 0.555 | 0.762 | 0.855 | — | |
| MixFormer-L2024.11 | 0.555 | 0.762 | 0.855 | — | |
| AOT-Bvariant=Base2022.10 | 0.541 | — | — | 0.533 | |
| MixFormer-22kPre-trained=ImageNet-22k, Mask prediction=Alpha-Refine [58]2022.03 | 0.535 | 0.761 | 0.854 | — | |
| MixCvT-22kPre-training=ImageNet-22k2023.02 | 0.535 | 0.761 | 0.854 | — | |
| Mixformer2024.01 | 0.535 | 0.761 | 0.854 | — | |
| STARK ST101+ARBackbone=ResNet-101, Refinement=Alpha-Refine2022.03 | 0.53 | 0.7 | 0.869 | — | |
| RPTPre-trained=ImageNet-22k2022.03 | 0.53 | 0.7 | 0.869 | — | |
| AiATrack2022.07 | 0.53 | 0.764 | 0.827 | — | |
| RPT2022.10 | 0.53 | — | — | 0.29 | |
| RPT2023.02 | 0.53 | 0.7 | 0.869 | — | |
| RPT2024.11 | 0.53 | 0.7 | 0.869 | — | |
| SBTvariant=large, mask generation=AlphaRefine [53]2022.03 | 0.529 | 0.753 | 0.834 | — | |
| DualTFRPre-trained=ImageNet-22k2022.03 | 0.528 | 0.755 | 0.836 | — | |
| DualTFR2022.07 | 0.528 | 0.755 | 0.836 | — | |
| DualTFR2023.02 | 0.528 | 0.755 | 0.836 | — | |
| MixViTPre-training=ConvMAE2023.02 | 0.528 | 0.734 | 0.857 | — | |
| MixFormer-1kPre-trained=ImageNet-1k, Mask prediction=Alpha-Refine [58]2022.03 | 0.527 | 0.746 | 0.833 | — | |
| MixCvT-1kPre-training=ImageNet-1k2023.02 | 0.527 | 0.746 | 0.833 | — | |
| OSTrack-384online template update mechanism=false, resolution=384x3842022.03 | 0.524 | 0.767 | 0.816 | — | |
| SeqTrack-B2024.01 | 0.522 | — | — | — | |
| OSTrack-256online template update mechanism=false, resolution=256x2562022.03 | 0.518 | 0.762 | 0.814 | — | |
| MixViT2023.02 | 0.518 | 0.728 | 0.841 | — | |
| SBTvariant=base, mask generation=AlphaRefine [53]2022.03 | 0.515 | 0.752 | 0.825 | — | |
| SBT2024.01 | 0.515 | 0.752 | 0.825 | — | |
| AOT-Svariant=Small2022.10 | 0.512 | — | — | 0.499 | |
| ToMP-101+ARBackbone=ResNet-101, Refinement=Alpha-Refine2022.03 | 0.507 | 0.75 | 0.798 | — | |
| STARK-ST50+ARbackbone=ResNet-50, mask prediction=true, Alpha-Refine (AR)=true, spatio-temporal=true2021.03 | 0.505 | 0.759 | 0.817 | — | |
| STARK ST50+ARBackbone=ResNet-50, Refinement=Alpha-Refine2022.03 | 0.505 | 0.759 | 0.817 | — | |
| STARKPre-trained=ImageNet-22k2022.03 | 0.505 | 0.759 | 0.817 | — | |
| STARKBackbone=ST502022.07 | 0.505 | 0.759 | 0.817 | — | |
| STARK-ST50online template update mechanism=true2022.03 | 0.505 | 0.759 | 0.817 | — | |
| STARK2023.02 | 0.505 | 0.759 | 0.817 | — | |
| STARK2024.01 | 0.505 | 0.759 | 0.819 | — | |
| STARK-ST101+ARbackbone=ResNet-101, mask prediction=true, Alpha-Refine (AR)=true, spatio-temporal=true2021.03 | 0.497 | 0.763 | 0.789 | — | |
| Starkvariant=st1012022.03 | 0.497 | 0.763 | 0.789 | — | |
| Ocean Plus2022.03 | 0.497 | 0.763 | 0.789 | — | |
| STARK-ST101online template update mechanism=true2022.03 | 0.497 | 0.763 | 0.789 | — | |
| ToMP2023.02 | 0.497 | 0.75 | 0.798 | — | |
| ToMP-50+ARBackbone=ResNet-50, Refinement=Alpha-Refine2022.03 | 0.496 | 0.754 | 0.793 | — | |
| OceanPlusmask prediction=true2021.03 | 0.491 | 0.685 | 0.842 | — | |
| Alpha Refine2022.03 | 0.491 | 0.685 | 0.842 | — | |
| OceanPlusPre-trained=ImageNet-22k2022.03 | 0.491 | 0.685 | 0.842 | — | |
| OceanPlus2022.07 | 0.491 | 0.685 | 0.842 | — | |
| OceanPlus2023.02 | 0.491 | 0.685 | 0.842 | — | |
| OceanPlus2024.11 | 0.491 | 0.685 | 0.842 | — | |
| AlphaRefmask prediction=true2021.03 | 0.482 | 0.754 | 0.777 | — | |
| AlphaRef2022.03 | 0.482 | 0.754 | 0.777 | — | |
| AFOD2022.03 | 0.482 | 0.754 | 0.777 | — | |
| AlphaRefPre-trained=ImageNet-22k2022.03 | 0.482 | 0.754 | 0.777 | — | |
| Alpha-Refine2022.07 | 0.482 | 0.754 | 0.777 | — | |
| AlphaRef2022.10 | 0.482 | — | — | 0.486 | |
| AlphaRefonline template update mechanism=true2022.03 | 0.482 | 0.754 | 0.777 | — | |
| AlphaRef2023.02 | 0.482 | 0.754 | 0.777 | — | |
| AlphaRef2024.01 | 0.482 | 0.754 | 0.777 | — | |
| AlphaRef2024.11 | 0.482 | 0.754 | 0.777 | — | |
| SBTvariant=small, mask generation=AlphaRefine [53]2022.03 | 0.477 | 0.75 | 0.775 | — | |
| LWTL2022.03 | 0.472 | 0.713 | 0.795 | — | |
| DeAOT-Tvariant=Tiny2022.10 | 0.472 | — | — | 0.463 | |
| AFOD2024.11 | 0.472 | 0.713 | 0.795 | — | |
| Ocean Fast2022.03 | 0.463 | 0.719 | 0.798 | — | |
| STARK-S50+ARbackbone=ResNet-50, mask prediction=true, Alpha-Refine (AR)=true, spatio-temporal=false2021.03 | 0.462 | 0.761 | 0.749 | — | |
| Starkvariant=s502022.03 | 0.462 | 0.761 | 0.749 | — | |
| STARK-S50online template update mechanism=false2022.03 | 0.462 | 0.761 | 0.749 | — | |
| FastOceanmask prediction=true2021.03 | 0.461 | 0.693 | 0.803 | — | |
| Ocean2022.03 | 0.461 | 0.693 | 0.803 | — | |
| DET502022.03 | 0.441 | 0.679 | 0.787 | — | |
| D3Smask prediction=true2021.03 | 0.439 | 0.699 | 0.769 | — | |
| D3SPre-trained=ImageNet-22k2022.03 | 0.439 | 0.699 | 0.769 | — | |
| D3Sonline template update mechanism=true2022.03 | 0.439 | 0.699 | 0.769 | — | |
| D3S2023.02 | 0.439 | 0.699 | 0.769 | — | |
| D3S2024.01 | 0.439 | 0.699 | 0.769 | — | |
| AOT-Tvariant=Tiny2022.10 | 0.435 | — | — | 0.433 | |
| Oceanmask prediction=true2021.03 | 0.43 | 0.693 | 0.754 | — | |
| Ocean2022.03 | 0.43 | 0.693 | 0.754 | — | |
| Oceanonline template update mechanism=true2022.03 | 0.43 | 0.693 | 0.754 | — | |
| Ocean2024.01 | 0.43 | 0.693 | 0.754 | — | |
| SBTvariant=light, mask generation=AlphaRefine [53]2022.03 | 0.415 | 0.742 | 0.712 | — | |
| SiamMarginmask prediction=true2021.03 | 0.356 | 0.698 | 0.64 | — | |
| SiamMaskmask prediction=true2021.03 | 0.321 | 0.624 | 0.648 | — | |
| SiamMask2022.03 | 0.321 | 0.624 | 0.648 | — |