Re-identification on RGBNT201
82.4mAPNEXT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| NEXTVenue=Ours, Backbone=CLIP-B/16, Modality=Multi-Modal2025.05 | 82.4 | 86.6 | 92 | 94.7 | |
| NEXTParams (M)=94.8, FLOPs (G)=44.8, Throughput (Images/s)=275.1, Use CLIP text encoder=true2025.05 | 82.4 | 86.6 | — | — | |
| FUSEModality setting=Multi2026.06 | 81.4 | 86.1 | 91.5 | 93.8 | |
| STMIBackbone Architecture=CNN, CLIP-based=true2026.02 | 81.2 | 83.4 | 90.2 | 91.6 | |
| IDEABackbone Architecture=CNN, CLIP-based=true2026.02 | 80.2 | 82.1 | 90 | 93.3 | |
| IDEAVenue=CVPR’25, Backbone=CLIP-B/16, Modality=Multi-Modal2025.05 | 80.2 | 82.1 | 90 | 93.3 | |
| IDEAParams (M)=91.7, FLOPs (G)=43.7, Throughput (Images/s)=299.5, Use CLIP text encoder=true2025.05 | 80.2 | 82.1 | — | — | |
| IDEAModality setting=Multi2026.06 | 80.2 | 82.1 | 90 | 93.3 | |
| NEXT (w/o Text)Params (M)=94.8, FLOPs (G)=36.0, Throughput (Images/s)=354.0, Use CLIP text encoder=false2025.05 | 79.2 | 85.5 | — | — | |
| DeMoBackbone Architecture=CNN, CLIP-based=true2026.02 | 79 | 82.3 | 88.8 | 92 | |
| DeMoVenue=AAAI’25, Backbone=CLIP-B/16, Modality=Multi-Modal2025.05 | 79 | 82.3 | 88.8 | 92 | |
| DeMoParams (M)=98.8, FLOPs (G)=35.1, Throughput (Images/s)=403.6, Use CLIP text encoder=false2025.05 | 79 | 82.3 | — | — | |
| DeMoModality setting=Multi2026.06 | 79 | 82.3 | 88.8 | 92 | |
| MambaProBackbone Architecture=CNN, CLIP-based=true2026.02 | 78.9 | 83.4 | 89.8 | 91.9 | |
| MambaProVenue=AAAI’25, Backbone=CLIP-B/16, Modality=Multi-Modal2025.05 | 78.9 | 83.4 | 89.8 | 91.9 | |
| MambaProParams (M)=74.8, FLOPs (G)=52.4, Throughput (Images/s)=243.2, Use CLIP text encoder=false2025.05 | 78.9 | 83.4 | — | — | |
| PromptMAVenue=T-IP’25, Backbone=CLIP-B/16, Modality=Multi-Modal2025.05 | 78.4 | 80.9 | 87 | 88.9 | |
| PromptMAParams (M)=107.9, FLOPs (G)=36.2, Throughput (Images/s)=343.5, Use CLIP text encoder=false2025.05 | 78.4 | 80.9 | — | — | |
| PromptMAModality setting=Multi2026.06 | 78.4 | 80.9 | 87 | 88.9 | |
| ICPL-ReIDVenue=T-MM’25, Backbone=CLIP-B/16, Modality=Multi-Modal2025.05 | 75.1 | 77.4 | 84.2 | 87.9 | |
| ICPL-ReIDParams (M)=45.4, FLOPs (G)=39.8, Throughput (Images/s)=358.4, Use CLIP text encoder=false2025.05 | 75.1 | 77.4 | — | — | |
| DESANetModality setting=Multi2026.06 | 74.6 | 77.6 | 87.1 | 91.3 | |
| TOP-ReIDBackbone Architecture=ViT, CLIP-based=false2026.02 | 72.3 | 76.6 | 84.7 | 89.4 | |
| TOP-ReIDVenue=AAAI’24, Backbone=ViT-B/16, Modality=Multi-Modal2025.05 | 72.3 | 76.6 | 84.7 | 89.4 | |
| TOP-ReIDParams (M)=324.5, FLOPs (G)=35.5, Throughput (Images/s)=398.9, Use CLIP text encoder=false2025.05 | 72.3 | 76.6 | — | — | |
| TOP-ReIDModality setting=Multi2026.06 | 72.3 | 76.6 | 84.7 | 89.4 | |
| HTTBackbone Architecture=ViT, CLIP-based=false2026.02 | 71.1 | 73.4 | 83.1 | 87.3 | |
| HTTVenue=AAAI’24, Backbone=ViT-B/16, Modality=Multi-Modal2025.05 | 71.1 | 73.4 | 83.1 | 87.3 | |
| HTTModality setting=Multi2026.06 | 71.1 | 73.4 | 83.1 | 87.3 | |
| RSCNetBackbone Architecture=ViT, CLIP-based=false2026.02 | 68.2 | 72.5 | — | — | |
| RSCNetModality setting=Multi2026.06 | 68.2 | 72.5 | — | — | |
| WTSF-ReIDBackbone Architecture=ViT, CLIP-based=false2026.02 | 67.9 | 72.2 | 83.4 | 89.7 | |
| WTSF-ReIDModality setting=Multi2026.06 | 67.9 | 72.2 | 83.4 | 89.7 | |
| EDITORBackbone Architecture=ViT, CLIP-based=false2026.02 | 66.5 | 68.3 | 81.1 | 88.2 | |
| EDITORVenue=CVPR’24, Backbone=ViT-B/16, Modality=Multi-Modal2025.05 | 66.5 | 68.3 | 81.1 | 88.2 | |
| EDITORParams (M)=119.3, FLOPs (G)=40.8, Throughput (Images/s)=335.1, Use CLIP text encoder=false2025.05 | 66.5 | 68.3 | — | — | |
| EDITORModality setting=Multi2026.06 | 66.5 | 68.3 | 81.1 | 88.2 | |
| TransReIDVenue=ICCV’21, Backbone=ViT-B/16, Modality=Single-Modal2025.05 | 63.8 | 65.8 | 78.5 | 83.9 | |
| UniCatBackbone Architecture=ViT, CLIP-based=false2026.02 | 57 | 55.7 | — | — | |
| LRMMBackbone Architecture=CNN, CLIP-based=false2026.02 | 52.3 | 53.4 | 64.6 | 73.2 | |
| LRMMModality setting=Multi2026.06 | 52.3 | 53.4 | 64.6 | 73.2 | |
| IEEEBackbone Architecture=CNN, CLIP-based=false2026.02 | 47.5 | 44.4 | 57.1 | 63.6 | |
| IEEEModality setting=Multi2026.06 | 47.5 | 44.4 | 57.1 | 63.6 | |
| IEEEVenue=AAAI’22, Backbone=ResNet, Modality=Multi-Modal2025.05 | 46.4 | 47.1 | 58.5 | 64.2 | |
| DENetBackbone Architecture=CNN, CLIP-based=false2026.02 | 42.4 | 42.2 | 55.3 | 64.5 | |
| DENetModality setting=Multi2026.06 | 42.4 | 42.2 | 55.3 | 64.5 | |
| PFNetBackbone Architecture=CNN, CLIP-based=false2026.02 | 38.5 | 38.9 | 52 | 58.4 | |
| PFNetVenue=AAAI’21, Backbone=ResNet, Modality=Multi-Modal2025.05 | 38.5 | 38.9 | 52 | 58.4 | |
| PFNetModality setting=Multi2026.06 | 38.5 | 38.9 | 52 | 58.4 | |
| PCBModality setting=Single2026.06 | 32.8 | 28.1 | 37.4 | 46.9 | |
| HAMNetVenue=AAAI’20, Backbone=ResNet, Modality=Multi-Modal2025.05 | 27.7 | 26.3 | 41.5 | 51.7 | |
| HAMNetModality setting=Multi2026.06 | 27.7 | 26.3 | 41.5 | 51.7 | |
| CALVenue=ICCV’21, Backbone=ResNet, Modality=Single-Modal2025.05 | 27.6 | 24.3 | 36.5 | 45.7 | |
| CALModality setting=Single2026.06 | 27.6 | 24.3 | 36.5 | 45.7 | |
| MLFNVenue=CVPR’18, Backbone=ResNet, Modality=Single-Modal2025.05 | 26.1 | 24.2 | 35.9 | 44.1 | |
| MLFNModality setting=Single2026.06 | 26.1 | 24.2 | 35.9 | 44.1 | |
| OSNetVenue=ICCV’19, Backbone=ResNet, Modality=Single-Modal2025.05 | 25.4 | 22.3 | 35.1 | 44.7 | |
| OSNetModality setting=Single2026.06 | 25.4 | 22.3 | 35.1 | 44.7 | |
| MUDeepModality setting=Single2026.06 | 23.8 | 19.7 | 33.1 | 44.3 | |
| HACNNVenue=CVPR’18, Backbone=ResNet, Modality=Single-Modal2025.05 | 21.3 | 19 | 34.1 | 42.8 | |
| HACNNModality setting=Single2026.06 | 21.3 | 19 | 34.1 | 42.8 |