@inproceedings{e595dd2368b94ff892c7e4998d5a9187,
title = "Glasses: Adapter-Enhanced DINOv3 with Boundary-Aware Multi-scale Representations for Medical Image Segmentation",
abstract = "Accurate medical image segmentation requires simultaneous modeling of robust global semantics and precise anatomical boundaries. While Vision Transformers (ViTs) exhibit strong global representation capability, they often struggle to capture fine-grained local structures and boundary details, particularly in scenarios with low contrast and blurred boundaries. To address this limitation, we propose Glasses, a novel framework that equips pre-trained Transformers with a structure-enhanced visual adaptation mechanism for medical image segmentation. Specifically, Glasses employs a frozen DINOv3 as the backbone encoder, with only the terminal layers fine-tuned. We introduce a Multi-Scale Deformable Attention Adapter that performs cross-scale interaction between lightweight convolutional structural priors and Transformer tokens extracted from specific intermediate layers of DINOv3, thereby transforming single-scale token embeddings into hierarchical representations enriched with both local structural cues and global context. This design preserves the stability of pre-trained semantic representations while improving the modeling of local anatomical structures and scale variations. Furthermore, we introduce an independent Edge-Aware Convolutional Encoder that explicitly learns boundary-enriched multi-scale representations under nested boundary supervision, providing a strong source of fine-grained local structural information that complements the adapter-enhanced Transformer features during decoding. A U-Net-style decoder then progressively fuses the adapter-enhanced Transformer features and the boundary-enriched convolutional representations to recover high-resolution segmentation outputs. Extensive experiments on the Synapse and ISIC datasets demonstrate that Glasses achieves Dice coefficients of 83.52\% and 91.60\%, respectively, significantly outperforming state-of-the-art (SOTA) methods. Source code is available at https://github.com/friay/Glasses.",
keywords = "Boundary supervision, DINOv3, Segmentation",
author = "Jiahao Zhu and Hengfei Cui and Hanlin Yin and Geng Chen and Yong Xia",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2027.; 29th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2026 ; Conference date: 27-09-2026 Through 01-10-2026",
year = "2027",
doi = "10.1007/978-3-032-38085-2\_22",
language = "英语",
isbn = "9783032380845",
series = "Lecture Notes in Computer Science",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "229--239",
editor = "Guang Yang and Ehsan Adeli and \{de Bruijne\}, Marleen and \{de Bruijne\}, Marleen and Papiez, \{Bartlomiej W.\} and Stefanie Speidel and Pallavi Tiwari and Guoyan Zheng and Mohammad Yaqub and Qi Dou and Islem Rekik",
booktitle = "Medical Image Computing and Computer Assisted Intervention - MICCAI 2026 - 29th International Conference, 2026, Proceedings",
}