GuiDINO: Rethinking Vision Foundation Model in Medical Image Segmentation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918364012609536 |
|---|---|
| author | Liang, Zhuonan Guo, Wei Gan, Jie Song, Yaxuan Chen, Runnan Chang, Hang Cai, Weidong |
| author_facet | Liang, Zhuonan Guo, Wei Gan, Jie Song, Yaxuan Chen, Runnan Chang, Hang Cai, Weidong |
| contents | Foundation vision models are increasingly adopted in medical image analysis. Due to domain shift, these pretrained models misalign with medical image segmentation needs without being fully fine-tuned or lightly adapted. We introduce GuiDINO, a framework that repositions native foundation model to acting as a visual guidance generator for downstream segmentation. GuiDINO extracts visual feature representation from DINOv3 and converts them into a spatial guide mask via a lightweight TokenBook mechanism, which aggregates token-prototype similarities. This guide mask gates feature activations in multiple segmentation backbones, thereby injecting foundation-model priors while preserving the inductive biases and efficiency of medical dedicated architectures. Training relies on a guide supervision objective loss that aligns the guide mask to ground-truth regions, optionally augmented by a boundary-focused hinge loss to sharpen fine structures. GuiDINO also supports parameter-efficient adaptation through LoRA on the DINOv3 guide backbone. Across diverse medical datasets and nnUNet-style inference, GuiDINO consistently improves segmentation quality and boundary robustness, suggesting a practical alternative to fine-tuning and offering a new perspective on how foundation models can best serve medical vision. Code is available at https://github.com/Hi-FishU/GuiDINO |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01115 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GuiDINO: Rethinking Vision Foundation Model in Medical Image Segmentation Liang, Zhuonan Guo, Wei Gan, Jie Song, Yaxuan Chen, Runnan Chang, Hang Cai, Weidong Computer Vision and Pattern Recognition Foundation vision models are increasingly adopted in medical image analysis. Due to domain shift, these pretrained models misalign with medical image segmentation needs without being fully fine-tuned or lightly adapted. We introduce GuiDINO, a framework that repositions native foundation model to acting as a visual guidance generator for downstream segmentation. GuiDINO extracts visual feature representation from DINOv3 and converts them into a spatial guide mask via a lightweight TokenBook mechanism, which aggregates token-prototype similarities. This guide mask gates feature activations in multiple segmentation backbones, thereby injecting foundation-model priors while preserving the inductive biases and efficiency of medical dedicated architectures. Training relies on a guide supervision objective loss that aligns the guide mask to ground-truth regions, optionally augmented by a boundary-focused hinge loss to sharpen fine structures. GuiDINO also supports parameter-efficient adaptation through LoRA on the DINOv3 guide backbone. Across diverse medical datasets and nnUNet-style inference, GuiDINO consistently improves segmentation quality and boundary robustness, suggesting a practical alternative to fine-tuning and offering a new perspective on how foundation models can best serve medical vision. Code is available at https://github.com/Hi-FishU/GuiDINO |
| title | GuiDINO: Rethinking Vision Foundation Model in Medical Image Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.01115 |