This work tackles the cross-domain object detection problem which aims to generalize a pre-trained object detector to different domains (driving scenes) without labels. An uncertainty-aware and class-balanced domain adaptation method is proposed based on two motivations: 1) estimation and exploitation of model uncertainty in a new domain is critical for reliable domain adaptation; and 2) in domain adaptation the distribution alignment of two domains as well as the maintaining of category discriminability are both important. In particular, we compose a Bayesian CNN-based framework for uncertainty estimation in object detection. We propose an algorithm for generating uncertainty-aware pseudo-labels, which are then used in uncertainty-guided self-training and category-aware feature alignment. We further devise a scheme with class-balanced memory banks to address the long-tail distribution problem in category-aware feature alignment. Experiments on multiple cross-domain object detection benchmarks show that our proposed method achieves state-of-the-art performance.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Uncertainty-Aware and Class-Balanced Domain Adaptation for Object Detection in Driving Scenes


    Contributors:

    Published in:

    Publication date :

    2024-11-01


    Size :

    2855670 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Ret3D: Rethinking Object Relations for Efficient 3D Object Detection in Driving Scenes

    Wu, Y. / Zhang, D. / Zhang, L. et al. | BASE | 2022

    Free access

    Uncertainty-Aware Pseudo Labels for Domain Adaptation in Pedestrian Trajectory Prediction

    Poibrenski, Atanas / Nozarian, Farzad / Rezaeianaran, Farzaneh et al. | IEEE | 2023



    Robust 3D Object Detection Based on Point Feature Enhancement in Driving Scenes

    Chen, Renjie / Zhang, Dongbo / Liu, Qinrui et al. | IEEE | 2024


    Short-Term Traffic Prediction with Balanced Domain Adaptation

    Zhang, Kaihang / Li, Junyi / Zhou, Qishen et al. | TIBKAT | 2022