Semantic segmentation in driving scenarios is important for modern autonomous driving technology. While the existing methods have shown promising results in segmenting normal-condition images, their performance in adverse scenes remains unsatisfactory due to limited visual field and lack of annotation. To address this issue, we propose an unsupervised domain adaptation semantic segmentation method with the transformer architecture, namely ACSegFormer, for driving-scene adverse conditions, aiming at mining image features in visually restricted scenes. Three effective training strategies are proposed in ACSegFormer to learn the latent image context relations and to reduce the gaps between different domains: an entropy-based pseudo label correction scheme that refines the target domain predictions with the normal reference predictions, an optimal transport-based inter-domain alignment module that performs domain alignment on the outputs of transformer encoder, and a masked context learning module that enhances the model’s ability to perceive the missing information of target domain image. Our ACSegFormer has no additional training parameters on top of the existing transformer segmentation framework, which can be easily used for self-training-based unsupervised domain adaptation approaches. The experimental results show that our ACSegFormer achieves state-of-the-art performance on driving-scene segmentation benchmarks in adverse conditions, including Dark Zurich and ACDC. Codes and models are available at https://github.com/wenyyu/ACSegFormer.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Domain Adaptation Transformer for Unsupervised Driving-Scene Segmentation in Adverse Conditions


    Beteiligte:
    Liu, Wenyu (Autor:in) / Wang, Song (Autor:in) / Zhu, Jianke (Autor:in) / Xie, Xuansong (Autor:in) / Zhang, Lei (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    01.12.2024


    Format / Umfang :

    5060323 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Nighttime Road Scene Parsing by Unsupervised Domain Adaptation

    Song, Can / Wu, Jin / Zhu, Lei et al. | IEEE | 2022


    Unsupervised Domain Adaptation via Shared Content Representation for Semantic Segmentation

    Hubschneider, Christian / Birkenbach, Marius / Zollner, J. Marius | IEEE | 2021


    Continual Unsupervised Domain Adaptation for Semantic Segmentation by Online Frequency Domain Style Transfer

    Termohlen, Jan-Aike / Klingner, Marvin / Brettin, Leon J. et al. | IEEE | 2021


    Unsupervised scene segmentation using sparse coding context

    Liu, Y. C. / Chen, H. T. | British Library Online Contents | 2013


    Threshold-Adaptive Unsupervised Focal Loss for Domain Adaptation of Semantic Segmentation

    Yan, Weihao / Qian, Yeqiang / Wang, Chunxiang et al. | IEEE | 2023