Semantic Segmentation on Crack500
69Dice ScoreTversky
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| TverskyModel=Detectron2+DeepLabv3+, beta=0.62026.03 | 69 | 10.47 | 77 | |
| CEDiceModel=Detectron2+DeepLabv3+2026.03 | 68 | 10.61 | 76 | |
| CEDice+CEDiceModel=Detectron2+DeepLabv3+2026.03 | 68 | 9.32 | 76 | |
| SCNP-CEDiceModel=Detectron2+DeepLabv3+, Method Variant=CEDice applied to SCNP2026.03 | 66 | 9.48 | 73 | |
| clDiceModel=Detectron2+DeepLabv3+, iter=102026.03 | 65 | 9.98 | 80 | |
| SkelRecallModel=Detectron2+DeepLabv3+2026.03 | 65 | 11.78 | 77 | |
| TopoLossModel=Detectron2+DeepLabv3+2026.03 | 54 | 13.34 | 58 |