Saved in:
Bibliographic Details
Main Authors: Meng, Xinru, Sun, Han, Liu, Jiamei, Liu, Ningzhong, Zhou, Huiyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.16692
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909590026715136
author Meng, Xinru
Sun, Han
Liu, Jiamei
Liu, Ningzhong
Zhou, Huiyu
author_facet Meng, Xinru
Sun, Han
Liu, Jiamei
Liu, Ningzhong
Zhou, Huiyu
contents Source-free domain adaptation (SFDA), which involves adapting models without access to source data, is both demanding and challenging. Existing SFDA techniques typically rely on pseudo-labels generated from confidence levels, leading to negative transfer due to significant noise. To tackle this problem, Energy-Based Pseudo-Label Refining (EBPR) is proposed for SFDA. Pseudo-labels are created for all sample clusters according to their energy scores. Global and class energy thresholds are computed to selectively filter pseudo-labels. Furthermore, a contrastive learning strategy is introduced to filter difficult samples, aligning them with their augmented versions to learn more discriminative features. Our method is validated on the Office-31, Office-Home, and VisDA-C datasets, consistently finding that our model outperformed state-of-the-art methods.
format Preprint
id arxiv_https___arxiv_org_abs_2504_16692
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Energy-Based Pseudo-Label Refining for Source-free Domain Adaptation
Meng, Xinru
Sun, Han
Liu, Jiamei
Liu, Ningzhong
Zhou, Huiyu
Computer Vision and Pattern Recognition
Source-free domain adaptation (SFDA), which involves adapting models without access to source data, is both demanding and challenging. Existing SFDA techniques typically rely on pseudo-labels generated from confidence levels, leading to negative transfer due to significant noise. To tackle this problem, Energy-Based Pseudo-Label Refining (EBPR) is proposed for SFDA. Pseudo-labels are created for all sample clusters according to their energy scores. Global and class energy thresholds are computed to selectively filter pseudo-labels. Furthermore, a contrastive learning strategy is introduced to filter difficult samples, aligning them with their augmented versions to learn more discriminative features. Our method is validated on the Office-31, Office-Home, and VisDA-C datasets, consistently finding that our model outperformed state-of-the-art methods.
title Energy-Based Pseudo-Label Refining for Source-free Domain Adaptation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.16692