Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Huang, Yuyang, Chen, Yabo, Zhou, Junyu, Dai, Wenrui, Zhang, Xiaopeng, Zou, Junni, Xiong, Hongkai, Tian, Qi
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2510.25279
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918177750908928
author Huang, Yuyang
Chen, Yabo
Zhou, Junyu
Dai, Wenrui
Zhang, Xiaopeng
Zou, Junni
Xiong, Hongkai
Tian, Qi
author_facet Huang, Yuyang
Chen, Yabo
Zhou, Junyu
Dai, Wenrui
Zhang, Xiaopeng
Zou, Junni
Xiong, Hongkai
Tian, Qi
contents Source-free domain adaptation (SFDA) is a challenging task that tackles domain shifts using only a pre-trained source model and unlabeled target data. Existing SFDA methods are restricted by the fundamental limitation of source-target domain discrepancy. Non-generation SFDA methods suffer from unreliable pseudo-labels in challenging scenarios with large domain discrepancies, while generation-based SFDA methods are evidently degraded due to enlarged domain discrepancies in creating pseudo-source data. To address this limitation, we propose a novel generation-based framework named Diffusion-Driven Progressive Target Manipulation (DPTM) that leverages unlabeled target data as references to reliably generate and progressively refine a pseudo-target domain for SFDA. Specifically, we divide the target samples into a trust set and a non-trust set based on the reliability of pseudo-labels to sufficiently and reliably exploit their information. For samples from the non-trust set, we develop a manipulation strategy to semantically transform them into the newly assigned categories, while simultaneously maintaining them in the target distribution via a latent diffusion model. Furthermore, we design a progressive refinement mechanism that progressively reduces the domain discrepancy between the pseudo-target domain and the real target domain via iterative refinement. Experimental results demonstrate that DPTM outperforms existing methods by a large margin and achieves state-of-the-art performance on four prevailing SFDA benchmark datasets with different scales. Remarkably, DPTM can significantly enhance the performance by up to 18.6% in scenarios with large source-target gaps.
format Preprint
id arxiv_https___arxiv_org_abs_2510_25279
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Diffusion-Driven Progressive Target Manipulation for Source-Free Domain Adaptation
Huang, Yuyang
Chen, Yabo
Zhou, Junyu
Dai, Wenrui
Zhang, Xiaopeng
Zou, Junni
Xiong, Hongkai
Tian, Qi
Computer Vision and Pattern Recognition
Source-free domain adaptation (SFDA) is a challenging task that tackles domain shifts using only a pre-trained source model and unlabeled target data. Existing SFDA methods are restricted by the fundamental limitation of source-target domain discrepancy. Non-generation SFDA methods suffer from unreliable pseudo-labels in challenging scenarios with large domain discrepancies, while generation-based SFDA methods are evidently degraded due to enlarged domain discrepancies in creating pseudo-source data. To address this limitation, we propose a novel generation-based framework named Diffusion-Driven Progressive Target Manipulation (DPTM) that leverages unlabeled target data as references to reliably generate and progressively refine a pseudo-target domain for SFDA. Specifically, we divide the target samples into a trust set and a non-trust set based on the reliability of pseudo-labels to sufficiently and reliably exploit their information. For samples from the non-trust set, we develop a manipulation strategy to semantically transform them into the newly assigned categories, while simultaneously maintaining them in the target distribution via a latent diffusion model. Furthermore, we design a progressive refinement mechanism that progressively reduces the domain discrepancy between the pseudo-target domain and the real target domain via iterative refinement. Experimental results demonstrate that DPTM outperforms existing methods by a large margin and achieves state-of-the-art performance on four prevailing SFDA benchmark datasets with different scales. Remarkably, DPTM can significantly enhance the performance by up to 18.6% in scenarios with large source-target gaps.
title Diffusion-Driven Progressive Target Manipulation for Source-Free Domain Adaptation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.25279