Segmentation on Cell Nucleus Many-shot
82Total DiceCellpose (3D)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Cellpose (3D)Pre-training Strategy=Baselines, Backbone=Cellpose2026.05 | 82 | 82 | |
| UNetPre-training Strategy=Image-only pretraining, Backbone=UNet2026.05 | 82 | 96 | |
| SwinPre-training Strategy=Trained from scratch, Backbone=Swin2026.05 | 81 | 97 | |
| Swin (over)Pre-training Strategy=Image+text pretraining, Backbone=Swin, Method Variant=overfit variant2026.05 | 81 | 98 | |
| UNetPre-training Strategy=Image+text pretraining, Backbone=UNet2026.05 | 81 | 94 | |
| UNetPre-training Strategy=Trained from scratch, Backbone=UNet2026.05 | 80 | 88 | |
| SwinPre-training Strategy=Image-only pretraining, Backbone=Swin2026.05 | 80 | 94 | |
| SwinPre-training Strategy=Image+text pretraining, Backbone=Swin2026.05 | 80 | 99 | |
| Cellpose (2D)Pre-training Strategy=Baselines, Backbone=Cellpose2026.05 | 57 | 43 | |
| uSAM (l)Pre-training Strategy=Baselines, Backbone=uSAM, Method Variant=large2026.05 | 56 | 22 | |
| uSAM (b)Pre-training Strategy=Baselines, Backbone=uSAM, Method Variant=base2026.05 | 52 | 10 | |
| CellSeg3DPre-training Strategy=Baselines, Backbone=CellSeg3D2026.05 | 51 | 74 |