Model card
TransUNet.
Chen et al. (JHU)open-source105M paramsViT encoder + U-Net decoder
First Transformer-based U-Net for medical image segmentation. Hybrid CNN+ViT architecture. CVPR 2021.
§ 01 · Benchmarks
Every benchmark TransUNet has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | ACDC | Medical · Medical Image Segmentation | mean-dsc | 89.1% | #6 | 2021-02-08 | source ↗ |
| 02 | Synapse Multi-Organ CT | Medical · Medical Image Segmentation | mean-dsc | 77.5% | #10 | 2021-02-08 | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 03 · Papers
1 paper with results for TransUNet.
- 2021-02-08· Medical· 2 results
TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation
§ 05 · Sources & freshness
Where these numbers come from.
arxiv
2
results
2 of 2 rows marked verified.