Visual Attribution on ImageNet Ground Truth Class (val)
0.352Deletion ScoreScoreCAM
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| ScoreCAMClassifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.821s2026.04 | 0.352 | 0.5001 | 0.1481 | 0.4856 | 0.6595 | 0.1739 | 52.7301 | 11.0598 | |
| AblationCAMClassifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.742s2026.04 | 0.3117 | 0.5429 | 0.2313 | 0.4336 | 0.7023 | 0.2688 | 44.5182 | 10.652 | |
| GradCAM++Classifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.019s2026.04 | 0.2959 | 0.5533 | 0.2573 | 0.4258 | 0.7331 | 0.3074 | 45.5688 | 11.9073 | |
| HiResCAMClassifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.019s2026.04 | 0.2578 | 0.5771 | 0.3193 | 0.36 | 0.7415 | 0.3815 | 37.0658 | 13.1477 | |
| GradCAMClassifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.004s2026.04 | 0.2578 | 0.5771 | 0.3193 | 0.36 | 0.7415 | 0.3815 | 37.065 | 13.1467 | |
| AHAClassifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.005s, Refinement steps (T)=02026.04 | 0.1624 | 0.5467 | 0.3843 | 0.2191 | 0.6828 | 0.4637 | 21.2367 | 12.6424 | |
| AHA T = 3Classifier=ConvNeXt Small [41], Reference image modes=black, mean, blur, Runtime=0.313s, Refinement steps (T)=32026.04 | 0.1472 | 0.5613 | 0.4141 | 0.1957 | 0.7028 | 0.5071 | 19.7873 | 14.0637 |