Image Attribution On Vggface2
評価指標
Deletion AUC score (ArcFace ResNet-101)
Insertion AUC score (ArcFace ResNet-101)
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Deletion AUC score (ArcFace ResNet-101) | Insertion AUC score (ArcFace ResNet-101) |
|---|---|---|
| a-unified-approach-to-interpreting-model | 0.2034 | 0.6132 |
| grad-cam-visual-explanations-from-deep | 0.3103 | 0.4733 |
| making-sense-of-dependence-efficient-black | 0.1317 | 0.6694 |
| deep-inside-convolutional-networks | 0.1907 | 0.5612 |
| less-is-more-fewer-interpretable-region-via | 0.1304 | 0.6705 |
| why-should-i-trust-you-explaining-the | 0.2119 | 0.6185 |
| axiomatic-attribution-for-deep-networks | 0.0749 | 0.5399 |
| rise-randomized-input-sampling-for | 0.1375 | 0.6530 |