A powerful scene understanding can be achieved by combining the tasks of semantic segmentation and instance level recognition. Considering that these tasks are complementary, we propose a multi-objective fusion scheme which leverages the capabilities of each task: pixel level semantic segmentation performs well in background classification and delimiting foreground objects from background, while instance level segmentation excels in recognizing and classifying objects as a whole. We use a fully convolutional residual network together with a feature pyramid network in order to achieve both semantic segmentation and Mask R-CNN based instance level recognition. We introduce a novel heuristic fusion approach for panoptic segmentation. The instance and semantic segmentation output of the network is fused into a panoptic segmentation. This is achieved using object sub-category class and instance propagation guidance by object category class from semantic segmentation. The proposed solution achieves significant improvements in semantic object segmentation and object mask boundaries refinement at low computational costs.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Fusion Scheme for Semantic and Instance-level Segmentation


    Contributors:


    Publication date :

    2018-11-01


    Size :

    1555353 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    EFFICIENT INSTANCE AND SEMANTIC SEGMENTATION FOR AUTOMATED DRIVING

    Petrovai, Andra / Nedevschi, Sergiu | British Library Conference Proceedings | 2019


    Efficient Instance and Semantic Segmentation for Automated Driving

    Petrovai, Andra / Nedevschi, Sergiu | IEEE | 2019


    Joint Semantic-Instance Segmentation Method for Intelligent Transportation System

    Li, Yujie / Cai, Jintong / Zhou, Quan et al. | IEEE | 2023


    Exploiting the Joint Potential of Instance Segmentation and Semantic Segmentation in Autonomous Driving

    Usman, Mohammed / K, Thirumala Akash / Ahmed, Mohammed Riyaz et al. | IEEE | 2023


    ImDeeplabV3plus with instance selective whitening loss in domain generalization semantic segmentation

    Zhang, You / Chen, Houjin / Li, Yanfeng et al. | Wiley | 2023

    Free access