• Home
  • Search
  • Towards Source-Free Domain Adaptive Semantic Segmentation via Importance-Aware and Prototype-Contrast Learning
  • Open Access IconOpen Access
  • Cite Icon19
  • https://doi.org/10.1109/tiv.2024.3383157Copy DOI Icon

Towards Source-Free Domain Adaptive Semantic Segmentation via Importance-Aware and Prototype-Contrast Learning

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Domain adaptive semantic segmentation enables robust pixel- wise understanding in real-world driving scenes. Source-free domain adaptation, as a more practical technique, addresses the concerns of data privacy and storage limitations in typical unsupervised domain adaptation methods, making it especially relevant in the context of intelligent vehicles. It utilizes a well-trained source model and unlabeled target data to achieve adaptation in the target domain. However, in the absence of source data and target labels, current solutions cannot sufficiently reduce the impact of domain shift and fully leverage the information from the target data. In this paper, we propose an end-to-end source-free domain adaptation semantic segmentation method via <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Importance-Aware and Prototype-Contrast (IAPC)</i> learning. The proposed IAPC framework effectively extracts domain-invariant knowledge from the well-trained source model and learns domain-specific knowledge from the unlabeled target domain. Specifically, considering the problem of domain shift in the prediction of the target domain by the source model, we put forward an importance-aware mechanism for the biased target prediction probability distribution to extract domain-invariant knowledge from the source model. We further introduce a prototype-contrast strategy, which includes a prototype-symmetric cross-entropy loss and a prototype-enhanced cross-entropy loss, to learn target intra-domain knowledge without relying on labels. A comprehensive variety of experiments on two domain adaptive semantic segmentation benchmarks demonstrates that the proposed end-to-end IAPC solution outperforms existing state-of-the-art methods. The source code is publicly available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/yihong-97/Source-free-IAPC</uri> .

Similar Papers
  • Dissertation

Label-efficient visual recognition with transfer learning

  • Jan 01, 2025
  • Jingyi Zhang
  • Research Article
  • Citations29

Unsupervised domain adaptation for person re-identification with iterative soft clustering

  • Dec 02, 2020
  • Knowledge-Based Systems
  • Jean-Paul Ainam +3
  • Research Article
  • Citations130

Source free domain adaptation for medical image segmentation with fourier style mining.

  • Jul 01, 2022
  • Medical Image Analysis
  • Chen Yang +3
  • Research Article
  • Citations4

Consistency-guided Multi-Source-Free Domain Adaptation

  • Oct 25, 2024
  • Engineering Applications of Artificial Intelligence
  • Ziyi Liu +6
  • Research Article
  • Citations7

Two-stage adversarial learning based unsupervised domain adaptation for retinal OCT segmentation.

  • Mar 01, 2024
  • Medical Physics
  • Shengyong Diao +8
  • Research Article

Robust Source-Free Domain Adaptation From Non-Robust Source Models.

  • Jan 01, 2026
  • IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
  • Yao Xiao +4
  • Research Article
  • Citations9

Open-Set Graph Domain Adaptation via Separate Domain Alignment

  • Mar 24, 2024
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Yu Wang +3
  • Research Article
  • Citations15

A Curriculum-Style Self-Training Approach for Source-Free Semantic Segmentation.

  • Dec 01, 2024
  • IEEE transactions on pattern analysis and machine intelligence
  • Yuxi Wang +2
  • Research Article
  • Citations45

Universal multi-Source domain adaptation for image classification

  • Aug 14, 2021
  • Pattern Recognition
  • Yueming Yin +3
  • Research Article
  • Citations16

Overcoming learning bias via Prototypical Feature Compensation for source-free domain adaptation

  • Sep 17, 2024
  • Pattern Recognition
  • Zicheng Pan +3
  • Research Article
  • Citations18

Leveraging topology for domain adaptive road segmentation in satellite and aerial imagery

  • Nov 09, 2023
  • ISPRS Journal of Photogrammetry and Remote Sensing
  • Javed Iqbal +3
  • PDF
  • Research Article
  • Citations6

Unsupervised Domain Adaptation for Vertebrae Detection and Identification in 3D CT Volumes Using a Domain Sanity Loss

  • Aug 19, 2022
  • Journal of Imaging
  • Pascal Sager +3
  • Research Article

Uncertainty-Aware Domain Adaptation for ECG Classification.

  • Jul 01, 2025
  • Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference
  • Dawnlicity Charls +2
  • Conference Article
  • Citations3

Domain Adaptation and Super-Resolution Based Bi-Directional Semantic Segmentation Method for Remote Sensing Images

  • Jul 17, 2022
  • Min Liang +1
  • Dissertation

Towards robust sensing and recognition : from statistical learning to transfer learning

  • Jan 01, 2020
  • Jianfei Yang
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.