Accurate and robust object detection is critical for autonomous driving. Image-based detectors face difficulties caused by low visibility in adverse weather conditions. Thus, radar-camera fusion is of particular interest but presents challenges in optimally fusing heterogeneous data sources. To approach this issue, we propose two new radar preprocessing techniques to better align radar and camera data. In addition, we introduce a Multi-Task Cross-Modality Attention-Fusion Network (MCAF-Net) for object detection, which includes two new fusion blocks. These allow for exploiting information from the feature maps more comprehensively. The proposed algorithm jointly detects objects and segments free space, which guides the model to focus on the more relevant part of the scene, namely, the occupied space. Our approach outperforms current state-of-the-art radar-camera fusion-based object detectors in the nuScenes dataset and achieves more robust results in adverse weather conditions and nighttime scenarios.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Multi-Task Cross-Modality Attention-Fusion for 2D Object Detection


    Contributors:


    Publication date :

    2023-09-24


    Size :

    4533552 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    AMMF: Attention-Based Multi-Phase Multi-Task Fusion for Small Contour Object 3D Detection

    Xie, Bangquan / Yang, Zongming / Yang, Liang et al. | IEEE | 2023


    A 3D Object Detection Method Based on Cross-Attention Multimodal Feature Fusion

    Tian, Juanxiu / Zhou, Yanping / Peng, Meng et al. | IEEE | 2024


    Machine Learning Multi-Modality Fusion Approaches Outperform Single-Modality & Traditional Approaches

    Garagic, Denis / Pelgrift, Daniel / Peskoe, Jacob et al. | IEEE | 2021



    Fusion of multi-modality volumetric medical imagery

    Aguilar, M. / New, J.R. | IEEE | 2002