Depth from Defocus on DDFF
0.0003MSEFOSSA
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| FOSSABackbone Architecture=ViT-B, Fine-tuning Status=finetuned, Evaluation Protocol=Trained on DDFF2026.03 | 0.0003 | 0.0148 | 11 | 0.25 | 93.2 | 98.7 | 99.4 | |
| FOSSABackbone Architecture=ViT-S, Fine-tuning Status=finetuned, Evaluation Protocol=Trained on DDFF2026.03 | 0.0004 | 0.0183 | 11 | 0.45 | 93.6 | 98.3 | 99.1 | |
| DualFocusEvaluation Protocol=Trained on DDFF2026.03 | 0.0005 | 0.0194 | 16 | 0.56 | 80 | 95.4 | 98.2 | |
| HybridDepthEvaluation Protocol=Trained on DDFF2026.03 | 0.0005 | 0.02 | 17 | 0.6 | 78.9 | 94.7 | 98.1 | |
| DFF-DFVEvaluation Protocol=Trained on DDFF2026.03 | 0.0006 | 0.0213 | 17 | 0.63 | 76.7 | 94.2 | 98.1 | |
| DefocusNetEvaluation Protocol=Trained on DDFF2026.03 | 0.0009 | 0.0255 | 17 | 0.6 | 72.6 | 94.2 | 97.9 | |
| DDFFEvaluation Protocol=Trained on DDFF2026.03 | 0.0009 | 0.0276 | 24 | 0.95 | 61.3 | 88.7 | 96.5 | |
| FOSSABackbone Architecture=ViT-B, Fine-tuning Status=pretrained, Evaluation Protocol=Zero-Shot2026.03 | 0.0013 | 0.0324 | 21 | 1.07 | 60.8 | 92.1 | 96.8 | |
| FOSSABackbone Architecture=ViT-S, Fine-tuning Status=pretrained, Evaluation Protocol=Zero-Shot2026.03 | 0.0015 | 0.0352 | 27 | 1.19 | 34.6 | 81.2 | 95.4 |