Predictive Uncertainty during model training can be used to assess wether a sample is correctly annotated or not. To see if this is also possible on remote sensing data, we applied the method on a building segmentation task.


    Access

    Download


    Export, share and cite



    Title :

    Using Predictive Uncertainty for Cleaning Noisy Annotations


    Contributors:

    Conference:

    2022 ; Jena, Deutschland


    Publication date :

    2022-11-01


    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Ensemble-Based Semi-Supervised Learning to Improve Noisy Soiling Annotations in Autonomous Driving

    Uricar, Michal / Sistu, Ganesh / Yahiaoui, Lucie et al. | IEEE | 2021


    Data-Driven Robust Predictive Control for Mixed Vehicle Platoons Using Noisy Measurement

    Lan, Jianglin / Zhao, Dezong / Tian, Daxin | IEEE | 2023

    Free access

    Towards sparse sensor annotations: Uncertainty-based active transfer learning for airfoil flow field prediction

    ZHANG, Yunyang / ZHENG, Xiaohu / GONG, Zhiqiang et al. | Elsevier | 2024

    Free access

    Document sharing with annotations

    SAHGAL ABHINANDAN / HOU JUSTIN ALEXANDER CHI-YOUNG / GUPTA GAURAV et al. | European Patent Office | 2023

    Free access

    Model Predictive Capability Assessment under Uncertainty

    Mahadevan, Sankaran / Rebba, Ramesh | AIAA | 2005