List your company

Visual Object Tracking on GOT-10k

84.7AOProContEXT
75.65278.00180.3582.699Oct 27, 2022Jun 7, 2023Jan 17, 2024Aug 28, 2024Apr 9, 2025Nov 19, 2025Jul 1, 2026

Evaluation results

MethodLinks
ProContEXTPre-training=MAE-ImageNet-1k, Dynamic Template=true, Extra Branch=false2022.1084.7—74.672.9—
SUMO-LSource=Ours, Model Size=L2026.0683.5—95.583.9—
OSTrackPre-training=MAE-ImageNet-1k, Dynamic Template=false, Extra Branch=false2022.1083.2—73.770.8—
ProContEXTPre-training=CLS-ImageNet-1k, Dynamic Template=true, Extra Branch=false2022.1082.7—7268—
SAMURAI-LSource=ArXiv’25, Model Size=L2026.0682.7—94.583.4—
ProContEXTPre-training=CLS-ImageNet-22k, Dynamic Template=true, Extra Branch=false2022.1082.6—72.569.3—
SUMO-SSource=Ours, Model Size=S2026.0682.5—93.281.2—
SUMO-BSource=Ours, Model Size=B2026.0682—9380.2—
RELO-L256Model Scale=Large, Input Resolution=2562026.0581.8—91.183.5—
SAMURAIYear=20262026.0281.7————
SAMURAIYear=20262026.0281.7————
SAMURAIYear=2026, Variant=Highest-resolution (ViT-L prioritized)2026.0781.7————
SUTrackResolution=384, Model Size=Large2024.1281.5—89.583.3—
SUTrack-LPublication=[AAAI’25]2026.0381.5————
SUTrack-L384Multi Modal=true2026.0581.5—89.583.3—
ARPTrack-L384Multi Modal=false2026.0581.5—90.680.5—
ARPTrack-L384Model Scale=Large, Input Resolution=3842026.0581.5—90.680.5—
ARPTrack-384Semantic Feature=ViT-ARP-L, Training Data of Tracker=Base+VastTrack+K700, Frames=7, Trainable Parameters=405M2026.0781.5——80.5—
OneTrackerV2-L384Multi Modal=true2026.0581.3—91.883.9—
SAM2LongModel=SAM2.1-L2024.1081.1————
SUTrackResolution=224, Model Size=Large2024.1281—90.482.4—
SUTrack-L224Multi Modal=true2026.0581—90.482.4—
SUTrack-L224Model Scale=Large, Input Resolution=2242026.0581—90.482.4—
SAM 2Model=SAM2.1-L2024.1080.7————
OneTrackerV2-L224Multi Modal=true2026.0580.5—90.282.7—
RELO-B256Model Scale=Base, Input Resolution=2562026.0580.5—90.581.2—
MITS2024.0380.4—89.775.9—
GOT-Edit-378Semantic Feature=DINOv2-L, Geometry Feature=VGGT, Training Data of Tracker=Base+VastTrack, Frames=3, Trainable Parameters=53M2026.0780.2——79.8—
MixFormerPre-training=CLS-ImageNet-22k, Dynamic Template=true, Extra Branch=true2022.1080—70.767.8—
SPMTrack-L384Multi Modal=false2026.0580—89.479.9—
MCITrack-384Semantic Feature=Fast-iTPN-L, Training Data of Tracker=Base+VastTrack, Frames=5, Trainable Parameters=287M2026.0780——80.2—
MixFormerPre-training=CLS-ImageNet-1k, Dynamic Template=true, Extra Branch=true2022.1079.9—71.265.8—
GOT-JEPAYear=20252026.0279.6————
GOT-JEPAGOT-10k Guidelines=true2026.0279.6————
OneTrackerV2-B384Multi Modal=true2026.0579.6—89.481.3—
GOT-JEPAVariant=Highest-resolution (ViT-L prioritized)2026.0779.6————
ARTrackV2Resolution=384, Model Size=Large2024.1279.5—87.879.6—
ARTrackv2Year=20242026.0279.5————
ARTrackv2Year=20242026.0279.5————
ARTrackV2-L384Multi Modal=false2026.0579.5—87.879.6—
ARTrackV2-L384Model Scale=Large, Input Resolution=3842026.0579.5—87.879.6—
GOT-Edit-378Semantic Feature=DINOv2-L, Geometry Feature=DA3-L, Training Data of Tracker=Base+VastTrack, Frames=3, Trainable Parameters=28M2026.0779.5——79.6—
ARTrackv2Year=2024, Variant=Highest-resolution (ViT-L prioritized)2026.0779.5————
SUTrackResolution=384, Model Size=Base2024.1279.3—8880—
SUTrack-BPublication=[AAAI’25]2026.0379.3————
SUTrack-B384Multi Modal=true2026.0579.3—8880—
SimTrackPre-training=CLIP-WIT, Dynamic Template=false, Extra Branch=false2022.1078.9—68.662.4—
ARTrackResolution=384, Model Size=Large2024.1278.5—87.477.8—
ARTrackModel size=Large2024.0378.5—87.477.8—
OneTrackerV2-B224Multi Modal=true2026.0578.4—87.879.6—
ODTrack2024.1078.2————
LORAT2024.1078.2————
ODTrackResolution=384, Model Size=Large2024.1278.2—87.277.3—
ODTrack-LPublication=[AAAI’24]2026.0378.2————
ODTrack-L384Multi Modal=false2026.0578.2—87.277.3—
ODTrack-L384Model Scale=Large, Input Resolution=3842026.0578.2—87.277.3—
STARKPre-training=CLS-ImageNet-1k, Dynamic Template=true, Extra Branch=true2022.1078.1—68.864.1—
SUTrackResolution=224, Model Size=Base2024.1277.9—87.578.5—
DeTrackResolution=384, Training Data=GOT-10k only2025.0177.9—86.574.9—
RTrackerModel size=Large, Backbone=ViT-Large2024.0377.9—8776.9—
MCITrack-B224Evaluation Regime=Non-real-time, Speed (fps) GPU=34, Speed (fps) CPU=2, Speed (fps) AGX=62026.0377.9—88.276.8—
SUTrack-B224Evaluation Regime=Non-real-time, Speed (fps) GPU=55, Speed (fps) CPU=6, Speed (fps) AGX=132026.0377.9—87.578.5—
SUTrack-B224Multi Modal=true2026.0577.9—87.578.5—
MCITrackSource=AAAI25, Supervised=✓2026.0577.9—88.276.8—
SUTrack-B224Model Scale=Base, Input Resolution=2242026.0577.9—87.578.5—
DeTrack-384Semantic Feature=Denoising ViT, Training Data of Tracker=Base, Frames=32026.0777.9——74.9—
ARPTrack-B256Multi Modal=false2026.0577.7—87.374.3—
ARPTrackSource=CVPR25, Supervised=✓2026.0577.7—87.374.3—
ARPTrack-B256Model Scale=Base, Input Resolution=2562026.0577.7—87.374.3—
LORATResolution=378, Model Size=Large2024.1277.5—86.278.1—
LoRATYear=20242026.0277.5————
LoRATYear=20242026.0277.5————
LORAT-LPublication=[ECCV’24]2026.0377.5————
LoRAT-L378Multi Modal=false2026.0577.5—86.278.1—
ARTrackV2Source=CVPR24, Supervised=✓2026.0577.5—8675.5—
DreamTrackSource=CVPR25, Supervised=✓2026.0577.5—87.174.2—
LoRAT-378Semantic Feature=DINOv2-L, Training Data of Tracker=Base, Frames=3, Trainable Parameters=32M2026.0777.5——78.1—
ToMP-378 (Reproduced)Semantic Feature=DINOv2-L, Training Data of Tracker=Base+VastTrack, Frames=3, Trainable Parameters=25M2026.0777.5——75.8—
LoRATYear=2024, Variant=Highest-resolution (ViT-L prioritized)2026.0777.5————
HIPTrackSource=CVPR24, Supervised=✓2026.0577.4—8874.5—
DeTrackResolution=256, Training Data=GOT-10k only2025.0177.1—86.173.5—
ODTrackResolution=384, Model Size=Base2024.1277—87.975.1—
ODTrack-BPublication=[AAAI’24]2026.0377————
ODTrack-B384Multi Modal=false2026.0577—87.975.1—
ODTrackSource=AAAI24, Supervised=✓2026.0577—87.975.1—
ODTrack-B384Source=AAAI’24, Resolution=3842026.0677—87.975.1—
PiVOTYear=20252026.0276.9————
PiVOTYear=20252026.0276.9————
PiVOT-378Semantic Feature=DINOv2-L, Training Data of Tracker=Base, Frames=3, Trainable Parameters=34M2026.0776.9——75.5—
PiVOT-L-27Venue=-2026.0776.9—87.675.5—
PiVOTYear=2025, Variant=Highest-resolution (ViT-L prioritized)2026.0776.9————
TransTPre-training=CLS-ImageNet-1k, Dynamic Template=false, Extra Branch=false2022.1076.8—67.160.9—
EVPTrack-B384Source=AAAI’24, Resolution=3842026.0676.6—86.773.9—
SPMTrack-B384Multi Modal=false2026.0576.5—85.976.3—
GOT-Edit-252Semantic Feature=DINOv2-L, Geometry Feature=DA3-L, Training Data of Tracker=Base+VastTrack, Frames=3, Trainable Parameters=51M2026.0776.4——75.8—
LMTrackSource=AAAI25, Supervised=✓2026.0576.3—87.173.9—
MGTrackYear=20252026.0276.2————
MGTrackYear=20252026.0276.2————
MGTrackYear=2025, Variant=Highest-resolution (ViT-L prioritized)2026.0776.2————
AQATrack-BPublication=[CVPR’24]2026.0376————