Weakly Supervised Object Localization on ILSVRC
77.1MaxBoxAccTeD-Loc*
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| TeD-Loc*Similarity Scores=patch-anchor dot product2025.01 | 77.1 | 71.8 | 76.7 | — | — | — | |
| TeD-Loc2025.01 | 75.6 | 70 | 75.1 | — | — | — | |
| GenPrompt2025.01 | 75 | 65.2 | 73.4 | — | — | — | |
| SeCM2025.01 | 72.9 | 61.7 | 70.9 | — | — | — | |
| BAS2025.01 | 72 | 58.5 | 69 | — | — | — | |
| DA-WSOL2025.01 | 71.8 | 55.3 | — | — | — | — | |
| CLIP-ES (Pred)Mapping=Grad-CAM2025.01 | 71.2 | — | — | — | — | — | |
| SATBackbone=DeiT-S, Model Selection=CLIP2024.04 | 70.13 | — | — | — | — | 62.8 | |
| SATBackbone=DeiT-S, Model Selection=SS2024.04 | 70.13 | — | — | — | — | 62.8 | |
| SATBackbone=DeiT-S, Model Selection=GT2024.04 | 70.12 | — | — | — | — | 62.8 | |
| CATR2025.01 | 69.2 | 56.9 | 66.6 | — | — | — | |
| SCM2025.01 | 68.8 | 56.1 | 66.4 | — | — | — | |
| LCTR2025.01 | 68.7 | 56.1 | 65.8 | — | — | — | |
| C2AM2025.01 | 68.5 | 59.6 | 67.1 | — | — | — | |
| TS-CAM2025.01 | 67.7 | 53.4 | 64.3 | — | — | — | |
| SATBackbone=DeiT-S, Model Selection=RPN2024.04 | 67.08 | — | — | — | — | 58.55 | |
| TS-CAMBackbone=DeiT-S, Model Selection=GT2024.04 | 66.75 | — | — | — | — | 59.54 | |
| TS-CAMBackbone=DeiT-S, Model Selection=RPN2024.04 | 66.75 | — | — | — | — | 59.54 | |
| TS-CAMBackbone=DeiT-S, Model Selection=CLIP2024.04 | 66.75 | — | — | — | — | 59.54 | |
| PSOL2025.01 | 66.3 | 58 | 65 | — | — | — | |
| TS-CAMBackbone=DeiT-S, Model Selection=SS2024.04 | 66.17 | — | — | — | — | 59 | |
| ADLBackbone=ResNet50, Model Selection=RPN2024.04 | 65.97 | — | — | — | — | 58.37 | |
| ADLBackbone=ResNet50, Model Selection=SS2024.04 | 65.19 | — | — | — | — | 58.54 | |
| ADLBackbone=ResNet50, Model Selection=GT2024.04 | 65.11 | — | — | — | — | 58.46 | |
| ADLBackbone=ResNet50, Model Selection=CLIP2024.04 | 65.11 | — | — | — | — | 58.55 | |
| SATBackbone=DeiT-S, Model Selection=CL2024.04 | 64.94 | — | — | — | — | 56.09 | |
| CAMBackbone=ResNet50, Model Selection=GT2024.04 | 64.06 | — | — | — | — | 57.89 | |
| HaSBackbone=ResNet50, Model Selection=RPN2024.04 | 63.94 | — | — | — | — | 58.39 | |
| CAMBackbone=ResNet50, Model Selection=CLIP2024.04 | 63.9 | — | — | — | — | 57.68 | |
| HaSBackbone=ResNet50, Model Selection=GT2024.04 | 63.77 | — | — | — | — | 58.49 | |
| HaSBackbone=ResNet50, Model Selection=CLIP2024.04 | 63.77 | — | — | — | — | 58.49 | |
| CAMBackbone=ResNet50, Model Selection=RPN2024.04 | 63.6 | — | — | — | — | 57.42 | |
| CAMBackbone=ResNet50, Model Selection=SS2024.04 | 63.6 | — | — | — | — | 57.42 | |
| ACoLBackbone=ResNet50, Model Selection=CLIP2024.04 | 63.45 | — | — | — | — | 56.94 | |
| HaSBackbone=ResNet50, Model Selection=SS2024.04 | 63.3 | — | — | — | — | 57.32 | |
| ACoLBackbone=ResNet50, Model Selection=GT2024.04 | 62.93 | — | — | — | — | 56.39 | |
| ACoLBackbone=ResNet50, Model Selection=SS2024.04 | 62.92 | — | — | — | — | 56.39 | |
| ADLBackbone=ResNet50, Model Selection=CL2024.04 | 62.81 | — | — | — | — | 56.39 | |
| ACoLBackbone=ResNet50, Model Selection=RPN2024.04 | 62.75 | — | — | — | — | 56.32 | |
| ACoLBackbone=ResNet50, Model Selection=CL2024.04 | 61.98 | — | — | — | — | 55.62 | |
| SCMBackbone=DeiT-S, Model Selection=GT2024.04 | 61.76 | — | — | — | — | 54.55 | |
| SCMBackbone=DeiT-S, Model Selection=RPN2024.04 | 61.75 | — | — | — | — | 54.56 | |
| SCMBackbone=DeiT-S, Model Selection=CLIP2024.04 | 61.75 | — | — | — | — | 54.56 | |
| HaSBackbone=ResNet50, Model Selection=CL2024.04 | 61.69 | — | — | — | — | 56.81 | |
| CAMBackbone=ResNet50, Model Selection=CL2024.04 | 61.48 | — | — | — | — | 56.17 | |
| NL-CCAMBackbone=VGG-GAP, Model Selection=GT2024.04 | 60.63 | — | — | — | — | 54.98 | |
| NL-CCAMBackbone=VGG-GAP, Model Selection=RPN2024.04 | 60.63 | — | — | — | — | 54.98 | |
| NL-CCAMBackbone=VGG-GAP, Model Selection=CLIP2024.04 | 60.63 | — | — | — | — | 54.98 | |
| SCMBackbone=DeiT-S, Model Selection=SS2024.04 | 59.76 | — | — | — | — | 53.47 | |
| NL-CCAMBackbone=VGG-GAP, Model Selection=CL2024.04 | 58.42 | — | — | — | — | 51.59 | |
| SCMBackbone=DeiT-S, Model Selection=CL2024.04 | 57.92 | — | — | — | — | 52.13 | |
| TS-CAMBackbone=DeiT-S, Model Selection=CL2024.04 | 56.4 | — | — | — | — | 53.67 | |
| NL-CCAMBackbone=VGG-GAP, Model Selection=SS2024.04 | 52.72 | — | — | — | — | 49.44 | |
| GenPromptHardware=NVIDIA-A100 GPU, Batch size=1, Warm-up=20 epochs, Inference steps=1,0002025.01 | — | — | — | 1,133.35 | 272 | — | |
| TeD-LocHardware=NVIDIA-A100 GPU, Batch size=1, Warm-up=20 epochs, Inference steps=1,0002025.01 | — | — | — | 569.67 | 121 | — |