Instance segmentation for autonomous driving should have a simple process. For dense instance segmentation methods based on FCOS, although they can directly predict instance masks, they output many duplicate predictions. Therefore, they must use non-maximum suppression (NMS) to remove the duplicate predictions and improve the performance. These methods with NMS not only hinder end-to-end training but also make the whole process more complicated. In contrast to these works, we propose a NMS-free FCOS based instance segmentation (NFIS). NFIS achieves comparable performance with NMS. Besides, we propose a context-based encoding module (CBEM) to refine the feature maps. CBEM utilizes the local and global information to refine the parameters of dynamic convolution and improve the classification. The proposed network outperforms most instance segmentation networks on the street scene datasets Cityscapes and Indian Driving Dataset (IDD).


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    NFIS: A NMS-free FCOS Method for Instance Segmentation


    Contributors:
    Du, Shan (author) / Chen, Zhaotang (author) / Li, Lingxi (author) / Zhang, Hui (author) / Cao, Dongpu (author) / Chen, Long (author)


    Publication date :

    2021-09-19


    Size :

    581323 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Robust Pedestrian Detection in Challenging Environmental Conditions Using FCOS

    Gonthina, Nagamani / Kola, Shiva Kumar / Dabbeti, Praneeth et al. | IEEE | 2023



    INSTANCE SEGMENTATION AS IMAGE SEGMENTATION ANNOTATION

    Watanabe, Thomio / Wolf, Denis F. | British Library Conference Proceedings | 2019


    Instance Segmentation as Image Segmentation Annotation

    Watanabe, Thomio / Wolf, Denis F. | IEEE | 2019


    Instance segmentation imaging system

    PHILION JONAH / ZHAO YIBIAO | European Patent Office | 2020

    Free access