Unsupervised Domain Adaptation via Similarity-based Prototypes for Cross-Modality Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ye, Ziyu, Ju, Chen, Ma, Chaofan, Zhang, Xiaoyun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915572069957632
author Ye, Ziyu
Ju, Chen
Ma, Chaofan
Zhang, Xiaoyun
author_facet Ye, Ziyu
Ju, Chen
Ma, Chaofan
Zhang, Xiaoyun
contents Deep learning models have achieved great success on various vision challenges, but a well-trained model would face drastic performance degradation when applied to unseen data. Since the model is sensitive to domain shift, unsupervised domain adaptation attempts to reduce the domain gap and avoid costly annotation of unseen domains. This paper proposes a novel framework for cross-modality segmentation via similarity-based prototypes. In specific, we learn class-wise prototypes within an embedding space, then introduce a similarity constraint to make these prototypes representative for each semantic class while separable from different classes. Moreover, we use dictionaries to store prototypes extracted from different images, which prevents the class-missing problem and enables the contrastive learning of prototypes, and further improves performance. Extensive experiments show that our method achieves better results than other state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2510_20596
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unsupervised Domain Adaptation via Similarity-based Prototypes for Cross-Modality Segmentation
Ye, Ziyu
Ju, Chen
Ma, Chaofan
Zhang, Xiaoyun
Computer Vision and Pattern Recognition
Artificial Intelligence
Deep learning models have achieved great success on various vision challenges, but a well-trained model would face drastic performance degradation when applied to unseen data. Since the model is sensitive to domain shift, unsupervised domain adaptation attempts to reduce the domain gap and avoid costly annotation of unseen domains. This paper proposes a novel framework for cross-modality segmentation via similarity-based prototypes. In specific, we learn class-wise prototypes within an embedding space, then introduce a similarity constraint to make these prototypes representative for each semantic class while separable from different classes. Moreover, we use dictionaries to store prototypes extracted from different images, which prevents the class-missing problem and enables the contrastive learning of prototypes, and further improves performance. Extensive experiments show that our method achieves better results than other state-of-the-art methods.
title Unsupervised Domain Adaptation via Similarity-based Prototypes for Cross-Modality Segmentation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2510.20596