Person Re-identification on DukeMTMC-reID
95.7Rank-1 AccBaseline+InSTD
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Baseline+InSTDCategory=Group 4: spatial-temporal methods2021.07 | 95.7 | 89.1 | 97.2 | 98 | |
| st-ReIDApproach=Others2021.04 | 94.4 | 83.9 | 97.4 | 98.2 | |
| Baseline+st-ReIDCategory=Group 4: spatial-temporal methods2021.07 | 94.1 | 84.3 | 96.3 | 97.2 | |
| st-ReIDSupervision Type=Fully supervised, Backbone=CNN, Reference=AAAI192022.11 | 94 | 82.8 | 97 | 97.8 | |
| DMCL2025.12 | 93.5 | 87.6 | — | — | |
| FlipReIDBackbone=ResNeSt50, Inference mode=double images, Flipping loss=with, Re-ranking=true2021.05 | 93 | 90.7 | — | — | |
| VT-ReID2025.12 | 92.6 | 79.2 | — | — | |
| FlipReIDBackbone=ResNet50, Inference mode=double images, Flipping loss=with, Re-ranking=true2021.05 | 92.5 | 90.9 | — | — | |
| BPBreIDBackbone=HRNet-W32, Training scheme=Arbitrary/Heavier2022.11 | 92.4 | 84.2 | — | — | |
| TMGF (w/ GT)Supervision Type=Fully supervised, Backbone=Transformer, Reference=This work2022.11 | 92.3 | 83.1 | 96.4 | 97.4 | |
| Adaptive L2 RegularizationBackbone=ResNet152, Re-ranking=applied2020.07 | 92.2 | 90.7 | — | — | |
| LVLM-ReIDBackbone=LVLM2024.11 | 92.2 | 82.8 | — | — | |
| FlipReIDBackbone=IBN-ResNet50, Inference mode=double images, Flipping loss=with, Re-ranking=true2021.05 | 91.8 | 89.8 | — | — | |
| DAAF-BoTre-ranking=true2020.03 | 91.7 | 89.6 | — | — | |
| DiPSize=384, Stride=122022.12 | 91.7 | 85.2 | — | — | |
| BPBreIDBackbone=ResNet-50-ibn, Training scheme=Arbitrary/Heavier2022.11 | 91.7 | 81.3 | — | — | |
| PLR-OSNet2020.01 | 91.6 | 81.2 | — | — | |
| VA-reID2020.01 | 91.6 | 84.5 | 96.2 | — | |
| DiPSize=256, Stride=122022.12 | 91.6 | 84.6 | — | — | |
| MPNMethod Category=SPF-based2020.03 | 91.5 | 82 | — | — | |
| LDSTraining scheme=Arbitrary/Heavier2022.11 | 91.5 | 82.5 | — | — | |
| CLIP-SCGIBackbone=ViT, Resolution=256x1282024.10 | 91.3 | 81.6 | — | — | |
| PFDSupervision Type=Fully supervised, Backbone=Transformer, Reference=AAAI222022.11 | 91.2 | 83.2 | — | — | |
| DiPSize=256, Stride=162022.12 | 91.2 | 83.8 | — | — | |
| DiPSize=384, Stride=162022.12 | 91.2 | 84.5 | — | — | |
| PFDvariant=*2024.01 | 91.2 | 83.2 | — | — | |
| PFDTraining scheme=Arbitrary/Heavier2022.11 | 91.2 | 83.2 | — | — | |
| TransReIDBackbone=DeiT-B/16, Size=256x128, Sliding-window setting=true2021.02 | 91.1 | 81.9 | — | — | |
| TransReIDBackbone=DeiT-B/16, Size=384x128, Sliding-window setting=true2021.02 | 91.1 | 82.1 | — | — | |
| SSGRTraining scheme=Arbitrary/Heavier2022.11 | 91.1 | 81.3 | — | — | |
| Resnet50-ibn-aalignment method=cluster anchor, Input size=384x1922020.08 | 91 | 81.8 | — | — | |
| SCSNVenue=CVPR 2020, Backbone=ResNet502020.07 | 91 | 79 | — | — | |
| SCSNBackbone=CNN, Size=384x1282021.02 | 91 | 79 | — | — | |
| SCSNBackbone=ResNet502021.05 | 91 | 79 | — | — | |
| SCSN (4-stage)stages=42020.08 | 91 | 79 | — | — | |
| SCSNSize=384, Stride=162022.12 | 91 | 79 | — | — | |
| PromptSGBackbone=ViT, Resolution=256x1282024.10 | 91 | 81.6 | — | — | |
| SCSN2025.12 | 91 | 79 | — | — | |
| PromptSG2026.03 | 91 | 81.6 | — | — | |
| FlipReIDBackbone=ResNeSt50, Inference mode=double images, Flipping loss=with, Re-ranking=false2021.05 | 90.9 | 81.5 | — | — | |
| SRFR2024.01 | 90.9 | 82 | — | — | |
| CLIP-ReIDBackbone=ViT, Side Information Embedding (SIE)=true, Overlapping Patch (OLP)=true2022.11 | 90.8 | 83.1 | — | — | |
| TransReIDBackbone=ViT-B/16, Size=256x128, Sliding-window setting=true2021.02 | 90.7 | 82 | — | — | |
| TransReIDBackbone=ViT-B/16, Size=384x128, Sliding-window setting=true2021.02 | 90.7 | 82.6 | — | — | |
| FlipReIDBackbone=ResNeSt50, Inference mode=double images, Flipping loss=without, Re-ranking=false2021.05 | 90.7 | 81.2 | — | — | |
| MHSA-Net+Re-rankingK=8, Re-ranking=true2020.08 | 90.7 | 87.2 | — | — | |
| TransReIDSupervision Type=Fully supervised, Backbone=Transformer, Reference=ICCV212022.11 | 90.7 | 82.6 | — | — | |
| TransReIDSize=256, Stride=122022.12 | 90.7 | 82 | — | — | |
| TransReIDSize=384, Stride=122022.12 | 90.7 | 82.6 | — | — | |
| TransReIDvariant=*2024.01 | 90.7 | 82 | — | — | |
| TransReIDBackbone=ViT, Resolution=256x1282024.10 | 90.7 | 82 | — | — | |
| TransReIDBackbone=ViT, Side Information Embedding (SIE)=true, Overlapping Patch (OLP)=true2022.11 | 90.7 | 82 | — | — | |
| TransReIDBackbone=ViT, Side Information Embedding (SIE)=true, Overlapping Patch (OLP)=true, Input Resolution=> 256x1282022.11 | 90.7 | 82.6 | — | — | |
| TransReIDInput image size=256x1282026.03 | 90.7 | 82 | — | — | |
| NFormerBackbone=ABDNet2022.04 | 90.6 | 85.7 | — | — | |
| NFormerSize=256, Stride=162022.12 | 90.6 | 85.7 | — | — | |
| PFDvariant=standard2024.01 | 90.6 | 82.2 | — | — | |
| FlipReIDBackbone=ResNeSt50, Inference mode=single image, Flipping loss=with, Re-ranking=false2021.05 | 90.5 | 80.7 | — | — | |
| LtReIDBackbone=CNN, Resolution=> 256x1282024.10 | 90.5 | 78.9 | — | — | |
| LTReIDBackbone=CNN, Input Resolution=> 256x1282022.11 | 90.5 | 80.4 | — | — | |
| TransReIDBackbone=ViT-B/16, Size=384x1282021.02 | 90.4 | 81.8 | — | — | |
| TransReIDSize=384, Stride=162022.12 | 90.4 | 81.8 | — | — | |
| LAG-NetBackbone=ResNet-502021.08 | 90.4 | 81.6 | 96 | — | |
| APNet-CBackbone=ResNet-502021.08 | 90.4 | 81.5 | 95.6 | — | |
| BoTre-ranking=true2020.03 | 90.3 | 89.1 | — | — | |
| Resnet50-ibn-aalignment method=cluster anchor, Input size=standard2020.08 | 90.2 | 80.6 | — | — | |
| Adaptive L2 RegularizationBackbone=ResNet1522020.07 | 90.2 | 81 | — | — | |
| TransReIDBackbone=DeiT-B/16, Size=384x1282021.02 | 90.2 | 81.3 | — | — | |
| FlipReIDBackbone=ResNet50, Inference mode=double images, Flipping loss=without, Re-ranking=false2021.05 | 90.2 | 79.8 | — | — | |
| PromptSGBackbone=CNN, Resolution=256x1282024.10 | 90.2 | 80.4 | — | — | |
| TransReIDBackbone=DeiT-B/16, Size=256x1282021.02 | 90.1 | 81.2 | — | — | |
| SCSN (3-stage)stages=32020.08 | 90.1 | 79 | — | — | |
| AAformerBackbone=ViT, Resolution=> 256x1282024.10 | 90.1 | 80 | — | — | |
| SCSNBackbone=ResNet-502021.08 | 90.1 | 79 | — | — | |
| AAformerBackbone=ViT, Input Resolution=> 256x1282022.11 | 90.1 | 80 | — | — | |
| AAformerInput image size=> 256x1282026.03 | 90.1 | 80.9 | — | — | |
| AAformerBackbone=ViT2024.11 | 90.1 | 80 | — | — | |
| FlipReIDBackbone=ResNeSt50, Inference mode=single image, Flipping loss=without, Re-ranking=false2021.05 | 90 | 79.9 | — | — | |
| CLIP-ReIDBackbone=CNN, Resolution=256x1282024.10 | 90 | 80.7 | — | — | |
| CLIP-ReIDBackbone=ViT, Resolution=256x1282024.10 | 90 | 82.5 | — | — | |
| CLIP-ReIDBackbone=CNN2022.11 | 90 | 80.7 | — | — | |
| CLIP-ReIDBackbone=ViT2022.11 | 90 | 82.5 | — | — | |
| CLIP-ReID2026.03 | 90 | 82.5 | — | — | |
| CLIP-ReIDBackbone=ViT2024.11 | 90 | 82.5 | — | — | |
| BaselineBackbone=ResNeSt50, Inference mode=double images, Re-ranking=false2021.05 | 89.9 | 80.5 | — | — | |
| ALDERBackbone=CNN, Resolution=> 256x1282024.10 | 89.9 | 78.9 | — | — | |
| ALDERBackbone=CNN, Input Resolution=> 256x1282022.11 | 89.9 | 78.9 | — | — | |
| MUVA2026.03 | 89.9 | 80.1 | — | — | |
| Compact-ReIDType=Attention-based, Number of Parameters=6.4M2019.10 | 89.8 | 80.3 | — | — | |
| Resnet50-ibn-aInput size=384x1922020.08 | 89.7 | 80 | — | — | |
| GCPApproach=Others2021.04 | 89.7 | 78.6 | — | — | |
| Relation-NetBackbone=CNN2024.11 | 89.7 | 78.6 | — | — | |
| ISPBackbone=CNN, Size=256x1282021.02 | 89.6 | 80 | — | — | |
| TransReIDBackbone=ViT-B/16, Size=256x1282021.02 | 89.6 | 80.6 | — | — | |
| FastReIDBackbone=ResNet502021.05 | 89.6 | 79.8 | — | — | |
| FlipReIDBackbone=IBN-ResNet50, Inference mode=double images, Flipping loss=with, Re-ranking=false2021.05 | 89.6 | 79.8 | — | — | |
| BaselineBackbone=ResNeSt50, Inference mode=single image, Re-ranking=false2021.05 | 89.6 | 80 | — | — | |
| ISP2022.04 | 89.6 | 80 | — | — | |
| ISPSize=256, Stride=162022.12 | 89.6 | 80 | — | — | |
| TransReIDSize=256, Stride=162022.12 | 89.6 | 80.6 | — | — |