Due to the difficulty of replicating the real conditions during training, supervised algorithms for spacecraft pose estimation experience a drop in performance when trained on synthetic data and applied to real operational data. To address this issue, we propose a test-time adaptation approach that leverages the temporal redundancy between images acquired during close proximity operations. Our approach involves extracting features from sequential spacecraft images, estimating their poses, and then using this information to synthesize a reconstructed view. We establish a self-supervised learning objective by comparing the synthesized view with the actual one. During training, we supervise both pose estimation and image synthesis, while at test time, we optimize the self-supervised objective. In addition, we introduce a regularization loss to prevent solutions that are not consistent with the keypoint structure of the spacecraft.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Test-Time Adaptation for Keypoint-Based Spacecraft Pose Estimation Based on Predicted-View Synthesis




    Publication date :

    2024-10-01


    Size :

    6345110 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English





    Kalman-based continuous pose estimation network for spacecraft

    Zhou, Rui / Qi, Naiming / Jia, Shuanli et al. | SPIE | 2024


    Spacecraft Pose Estimation Dataset (SPEED)

    Posso, Julien | DataCite | 2024


    Spacecraft Pose Estimation Based on Unsupervised Domain Adaptation and on a 3D-Guided Loss Combination

    Pérez-Villar, Juan Ignacio Bravo / García-Martín, Álvaro / Bescós, Jesús | ArXiv | 2022

    Free access