This chapter presents two common perception interfaces for capturing the relevant motions during the demonstrations, including optical tracking systems and vision cameras. The optical tracking system employs a set of infrared optical markers for capturing the demonstrated motions. Based on the markers’ locations with respect to a fixed reference frame, poses of predefined rigid bodies are inferred over a set of discrete time instants. A Firefly MV CMOS camera from Point Grey has also been used for the observation of demonstrations. The task perception with vision cameras is performed via acquiring visual information from the environment in the form of sequences of images. While the optical tracking systems provide accurate and reliable means for capturing demonstrated tasks, the vision cameras allow natural and unrestrained motions during the demonstrations which can quickly be performed without the need for attaching markers on the objects in the scene.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Task Perception


    Contributors:


    Publication date :

    2017-03-10


    Size :

    5 pages




    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English




    Perception and decision in pedestrian task

    Older, S. J. / Grayson, G. B. | TIBKAT | 1974


    Is Driving an Automatic Task? Dual-Task Performance and Hazard Perception

    McKenna, F. P. / Crick, J. / Transport Research Laboratory | British Library Conference Proceedings | 1994


    MtpNet: Multi-Task Panoptic Driving Perception Network

    Li, Zheng / Yuan, Xiaohui / Sun, Bifan et al. | IEEE | 2025


    Integrated Sensory-Motor Systems and Task Specific Perception

    Trivedi, M. M. | British Library Conference Proceedings | 1994


    Task-adaptable, pervasive perception for robots performing everyday manipulation

    Bálint-Benczédi, Ferenc / Universität Bremen | TIBKAT | 2020

    Free access