Image-to-image retrieval on Food101
87.53mAPSPBH
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SPBH# bits=482026.07 | 87.53 | — | — | — | |
| SPBH# bits=322026.07 | 86.57 | — | — | — | |
| RFHNet# bits=322026.07 | 86.44 | — | — | — | |
| RFHNet# bits=482026.07 | 86.42 | — | — | — | |
| RFHNet# bits=242026.07 | 86.17 | — | — | — | |
| SPBH# bits=242026.07 | 86.07 | — | — | — | |
| RFHNet# bits=122026.07 | 85.14 | — | — | — | |
| A2-Net++# bits=482026.07 | 83.66 | — | — | — | |
| FoodHash# bits=482026.07 | 83.44 | — | — | — | |
| AMGH# bits=482026.07 | 83.21 | — | — | — | |
| DAHNet# bits=482026.07 | 83.11 | — | — | — | |
| FoodHash# bits=242026.07 | 83.01 | — | — | — | |
| A2-Net++# bits=322026.07 | 82.91 | — | — | — | |
| FoodHash# bits=322026.07 | 82.62 | — | — | — | |
| SEMICON# bits=482026.07 | 82.44 | — | — | — | |
| AMGH# bits=322026.07 | 82.31 | — | — | — | |
| A2-Net# bits=482026.07 | 82.13 | — | — | — | |
| DAHNet# bits=322026.07 | 82.05 | — | — | — | |
| A2-Net++# bits=242026.07 | 81.46 | — | — | — | |
| AMGH# bits=242026.07 | 80.94 | — | — | — | |
| SPBH# bits=122026.07 | 80.7 | — | — | — | |
| SEMICON# bits=322026.07 | 80.19 | — | — | — | |
| DAHNet# bits=242026.07 | 79.38 | — | — | — | |
| FoodHash# bits=122026.07 | 77.26 | — | — | — | |
| SEMICON# bits=242026.07 | 76.57 | — | — | — | |
| A2-Net# bits=322026.07 | 74.27 | — | — | — | |
| OriginalBackbone=DINOv3, ViT=L/16, Similarity=⟨I, I⟩2026.03 | 71 | — | — | — | |
| DAHNet# bits=122026.07 | 67.67 | — | — | — | |
| A2-Net# bits=242026.07 | 66.87 | — | — | — | |
| PCA←Backbone=SigLIP2, ViT=B/16, Similarity=⟨I←, I←⟩2026.03 | 65.3 | — | — | — | |
| PCA←Backbone=CLIP, ViT=L/14, Similarity=⟨I←, I←⟩2026.03 | 64.4 | — | — | — | |
| IsoCLIPIntra-modal=true, Backbone=ViT-L/14, Latency (ms)=11 ± 12026.03 | 62.9 | — | — | — | |
| AMGH# bits=122026.07 | 62.59 | — | — | — | |
| PCA←Backbone=SigLIP, ViT=B/16, Similarity=⟨I←, I←⟩2026.03 | 61.8 | — | — | — | |
| OriginalBackbone=SigLIP2, ViT=B/16, Similarity=⟨I, I⟩2026.03 | 59.7 | — | — | — | |
| OTIBackbone=SigLIP, ViT=B/16, Similarity=⟨T, I⟩2026.03 | 59 | — | — | — | |
| IsoCLIPIntra-modal=true, Backbone=PE-Core-B-16, Latency (ms)=8 ± 12026.03 | 57.6 | — | — | — | |
| OriginalBackbone=SigLIP, ViT=B/16, Similarity=⟨I, I⟩2026.03 | 56.3 | — | — | — | |
| OTIBackbone=CLIP, ViT=L/14, Similarity=⟨T, I⟩2026.03 | 55.9 | — | — | — | |
| OTI (I→T)Intra-modal=false, Backbone=ViT-L/14, Latency (ms)=1872 ± 912026.03 | 55.9 | — | — | — | |
| OTI (I→T)Intra-modal=false, Backbone=PE-Core-B-16, Latency (ms)=3252 ± 352026.03 | 55.8 | — | — | — | |
| OriginalBackbone=CLIP, ViT=L/14, Similarity=⟨I, I⟩2026.03 | 55 | — | — | — | |
| IsoCLIPIntra-modal=true, Backbone=ViT-B/16-open, Latency (ms)=6 ± 12026.03 | 54.7 | — | — | — | |
| A2-Net++# bits=122026.07 | 54.51 | — | — | — | |
| OTI (I→T)Intra-modal=false, Backbone=ViT-B/16-open, Latency (ms)=1836 ± 832026.03 | 54.4 | — | — | — | |
| IsoCLIPIntra-modal=true, Backbone=ViT-B/16, Latency (ms)=6 ± 12026.03 | 53.5 | — | — | — | |
| Image-ImageIntra-modal=true, Backbone=PE-Core-B-16, Latency (ms)=8 ± 12026.03 | 53.4 | — | — | — | |
| Image-ImageIntra-modal=true, Backbone=ViT-L/14, Latency (ms)=11 ± 12026.03 | 53 | — | — | — | |
| Image-ImageIntra-modal=true, Backbone=ViT-B/16-open, Latency (ms)=6 ± 12026.03 | 51.1 | — | — | — | |
| SEMICON# bits=122026.07 | 50 | — | — | — | |
| ADSH# bits=482026.07 | 48.81 | — | — | — | |
| OTI (I→T)Intra-modal=false, Backbone=ViT-B/16, Latency (ms)=1856 ± 562026.03 | 47.5 | — | — | — | |
| A2-Net# bits=122026.07 | 46.44 | — | — | — | |
| PCA←Backbone=CLIP, ViT=B/32, Similarity=⟨I←, I←⟩2026.03 | 43 | — | — | — | |
| ADSH# bits=322026.07 | 42.89 | — | — | — | |
| Image-ImageIntra-modal=true, Backbone=ViT-B/16, Latency (ms)=6 ± 12026.03 | 42.8 | — | — | — | |
| ADSH# bits=242026.07 | 40.93 | — | — | — | |
| HashNet# bits=482026.07 | 39.65 | — | — | — | |
| IsoCLIPIntra-modal=true, Backbone=ViT-B/32, Latency (ms)=7 ± 12026.03 | 38 | — | — | — | |
| HashNet# bits=322026.07 | 35.9 | — | — | — | |
| ADSH# bits=122026.07 | 35.64 | — | — | — | |
| OTIBackbone=CLIP, ViT=B/32, Similarity=⟨T, I⟩2026.03 | 34.7 | — | — | — | |
| OTI (I→T)Intra-modal=false, Backbone=ViT-B/32, Latency (ms)=1879 ± 352026.03 | 34.7 | — | — | — | |
| HashNet# bits=242026.07 | 34.48 | — | — | — | |
| OriginalBackbone=CLIP, ViT=B/32, Similarity=⟨I, I⟩2026.03 | 33.6 | — | — | — | |
| Image-ImageIntra-modal=true, Backbone=ViT-B/32, Latency (ms)=7 ± 12026.03 | 32.3 | — | — | — | |
| PCA←Backbone=SLIP, ViT=B/16, Similarity=⟨I←, I←⟩2026.03 | 25.1 | — | — | — | |
| HashNet# bits=122026.07 | 24.42 | — | — | — | |
| OTIBackbone=SLIP, ViT=B/16, Similarity=⟨T, I⟩2026.03 | 20 | — | — | — | |
| OriginalBackbone=SLIP, ViT=B/16, Similarity=⟨I, I⟩2026.03 | 19.5 | — | — | — | |
| CS3H# bits=96, Backbone=ViT-L/162026.05 | 17.02 | — | — | — | |
| CS3H# bits=96, Backbone=ResNet502026.05 | 16.84 | — | — | — | |
| CS3H# bits=96, Backbone=VGG162026.05 | 16.74 | — | — | — | |
| CS3H# bits=48, Backbone=ViT-L/162026.05 | 15.64 | — | — | — | |
| CS3H# bits=48, Backbone=ResNet502026.05 | 15.63 | — | — | — | |
| CS3H# bits=48, Backbone=VGG162026.05 | 15.32 | — | — | — | |
| CS3H# bits=32, Backbone=ViT-L/162026.05 | 15.21 | — | — | — | |
| CS3H# bits=32, Backbone=VGG162026.05 | 15.03 | — | — | — | |
| CS3H# bits=32, Backbone=ResNet502026.05 | 14.48 | — | — | — | |
| CS3H# bits=24, Backbone=ViT-L/162026.05 | 14.07 | — | — | — | |
| CS3H# bits=24, Backbone=VGG162026.05 | 13.86 | — | — | — | |
| CS3H# bits=24, Backbone=ResNet502026.05 | 13.12 | — | — | — | |
| SPQ# bits=96, Backbone=ResNet502026.05 | 12.46 | — | — | — | |
| SPQ# bits=48, Backbone=ResNet502026.05 | 10.98 | — | — | — | |
| SPQ# bits=24, Backbone=ResNet502026.05 | 10.81 | — | — | — | |
| CS3H# bits=12, Backbone=ResNet502026.05 | 9.84 | — | — | — | |
| CS3H# bits=12, Backbone=ViT-L/162026.05 | 9.41 | — | — | — | |
| SPQ# bits=32, Backbone=ResNet502026.05 | 9.35 | — | — | — | |
| CS3H# bits=12, Backbone=VGG162026.05 | 9.25 | — | — | — | |
| SGH# bits=962026.05 | 9.12 | — | — | — | |
| SGH# bits=482026.05 | 8.26 | — | — | — | |
| SGH# bits=322026.05 | 7.05 | — | — | — | |
| SPQ# bits=12, Backbone=ResNet502026.05 | 7.03 | — | — | — | |
| SGH# bits=242026.05 | 4.83 | — | — | — | |
| SGH# bits=122026.05 | 3.81 | — | — | — | |
| A2-SSL2026.05 | — | 8.5 | 14.1 | 15.4 | |
| Bihalf2026.05 | — | 6.2 | 9.2 | 10.2 | |
| CIBHash2026.05 | — | 7.3 | 11.7 | 12.2 | |
| HashSCD2026.05 | — | 9.2 | 15 | 15.6 | |
| MeCoQ2026.05 | — | 7.9 | 11.6 | 13.2 |