Dense object tracking, the ability to localize specific object points with pixel-level accuracy, is an important computer vision task with numerous downstream applications in robotics. Existing approaches either compute dense keypoint embeddings in a single forward pass, meaning the model is trained to track everything at once, or allocate their full capacity to a sparse predefined set of points, trading generality for accuracy. In this paper we explore a middle ground based on the observation that the number of relevant points at a given time are typically relatively few, e.g. grasp points on a target object. Our main contribution is a novel architecture, inspired by few-shot task adaptation, which allows a sparse-style network to condition on a keypoint embedding that indicates which point to track. Our central finding is that this approach provides the generality of dense-embedding models, while offering accuracy significantly closer to sparse-keypoint approaches. We present results illustrating this capacity vs. accuracy trade-off, and demonstrate the ability to zero-shot transfer to new object instances (within-class) using a real-robot pick-and-place task.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Few-Shot Keypoint Detection as Task Adaptation via Latent Embeddings


    Beteiligte:
    Vecerik, M (Autor:in) / Kay, J (Autor:in) / Hadsell, R (Autor:in) / Agapito, L (Autor:in) / Scholz, J (Autor:in)

    Erscheinungsdatum :

    2022-07-12


    Anmerkungen:

    In: Proceedings - IEEE International Conference on Robotics and Automation. (pp. pp. 1251-1257). IEEE: Philadelphia, PA, USA. (2022)


    Medientyp :

    Paper


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629



    MONOCULAR 2D SEMANTIC KEYPOINT DETECTION AND TRACKING

    CHEN HAOFENG / ARJUN BHARGAVA / RARES ANDREI AMBRUS et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    KEYPOINT BASED ACTION LOCALIZATION

    KADAV ASIM / LAI FARLEY / GRAF HANS PETER et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Performance Evaluation of 3D Keypoint Detectors

    Tombari, F. / Salti, S. / Di Stefano, L. | British Library Online Contents | 2013


    Representation Learning for Robot Keypoint Detection using Prior Kinematic Knowledge

    Klüpfel, Leonard | Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2022

    Freier Zugriff

    Keypoint Detection and Local Feature Matching for Textured 3D Face Recognition

    Mian, A. S. / Bennamoun, M. / Owens, R. | British Library Online Contents | 2008