Multi-Source Soft Pseudo-Label Learning with Domain Similarity-based Weighting for Semantic Segmentation

被引:2
作者
Matsuzaki, Shigemichi [1 ]
Masuzawa, Hiroaki [1 ]
Miura, Jun [1 ]
机构
[1] Toyohashi Univ Technol, Dept Comp Sci & Engn, Hibarigaoka 1-1,Tenpaku Cho, Toyohashi, Aichi, Japan
来源
2023 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) | 2023年
关键词
D O I
10.1109/IROS55552.2023.10342159
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper describes a method of domain adaptive training for semantic segmentation using multiple source datasets that are not necessarily relevant to the target dataset. We propose a soft pseudo-label generation method by integrating predicted object probabilities from multiple source models. The prediction of each source model is weighted based on the estimated domain similarity between the source and the target datasets to emphasize contribution of a model trained on a source that is more similar to the target and generate reasonable pseudo-labels. We also propose a training method using the soft pseudo-labels considering their entropy to fully exploit information from the source datasets while suppressing the influence of possibly misclassified pixels. The experiments show comparative or better performance than our previous work and another existing multi-source domain adaptation method, and applicability to a variety of target environments.
引用
收藏
页码:5852 / 5857
页数:6
相关论文
共 26 条
  • [1] A theory of learning from different domains
    Ben-David, Shai
    Blitzer, John
    Crammer, Koby
    Kulesza, Alex
    Pereira, Fernando
    Vaughan, Jennifer Wortman
    [J]. MACHINE LEARNING, 2010, 79 (1-2) : 151 - 175
  • [2] Semantic object classes in video: A high-definition ground truth database
    Brostow, Gabriel J.
    Fauqueur, Julien
    Cipolla, Roberto
    [J]. PATTERN RECOGNITION LETTERS, 2009, 30 (02) : 88 - 97
  • [3] The Cityscapes Dataset for Semantic Urban Scene Understanding
    Cordts, Marius
    Omran, Mohamed
    Ramos, Sebastian
    Rehfeld, Timo
    Enzweiler, Markus
    Benenson, Rodrigo
    Franke, Uwe
    Roth, Stefan
    Schiele, Bernt
    [J]. 2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, : 3213 - 3223
  • [4] Dosovitskiy A, 2017, P 1 ANN C ROB LEARN, P1, DOI [DOI 10.48550/ARXIV.1711.03938, 10.48550/arxiv.1711.03938]
  • [5] Gretton A, 2012, J MACH LEARN RES, V13, P723
  • [6] Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentation
    He, Jianzhong
    Jia, Xu
    Chen, Shuaijun
    Liu, Jianzhuang
    [J]. 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021, 2021, : 11003 - 11012
  • [7] Momentum Contrast for Unsupervised Visual Representation Learning
    He, Kaiming
    Fan, Haoqi
    Wu, Yuxin
    Xie, Saining
    Girshick, Ross
    [J]. 2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2020), 2020, : 9726 - 9735
  • [8] Hoffman J, 2018, PR MACH LEARN RES, V80
  • [9] Who is closer: A computational method for domain gap evaluation
    Liu, Xiaobin
    Zhang, Shiliang
    [J]. PATTERN RECOGNITION, 2022, 122
  • [10] Long M, 2016, PROCEEDINGS OF SYMPOSIUM OF POLICING DIPLOMACY AND THE BELT & ROAD INITIATIVE, 2016, P136