Object, scene and ego-centric action classification for robotic vision

Hasan Firdaus Bin Mohd Zaki

    Research output: ThesisDoctoral Thesis

    107 Downloads (Pure)

    Abstract

    Social robotics research focuses on developing intelligent machines that can perform difficult tasks especially those beyond human capabilities, involving repetitions or adverse conditions. However, robotic vision for high-level recognition associated with reasoning of uncertainty in unstructured environments is still far-fetched compared to the visual comprehension of humans. This dissertation investigates techniques to exploit the rich information from multi-modality sensors in pursuit of extending the frontiers of robotic vision. Three robot-centric recognition tasks are specifically explored; object, scene and action recognition with particular emphasis on designing highly effective and efficient feature representation and recognition algorithms.
    Original languageEnglish
    QualificationDoctor of Philosophy
    Awarding Institution
    • The University of Western Australia
    Award date1 Nov 2017
    DOIs
    Publication statusUnpublished - 2017

    Fingerprint

    Robotics
    Robots
    Sensors
    Uncertainty

    Cite this

    @phdthesis{f54cc3a5426d42d2a566e247ca491086,
    title = "Object, scene and ego-centric action classification for robotic vision",
    abstract = "Social robotics research focuses on developing intelligent machines that can perform difficult tasks especially those beyond human capabilities, involving repetitions or adverse conditions. However, robotic vision for high-level recognition associated with reasoning of uncertainty in unstructured environments is still far-fetched compared to the visual comprehension of humans. This dissertation investigates techniques to exploit the rich information from multi-modality sensors in pursuit of extending the frontiers of robotic vision. Three robot-centric recognition tasks are specifically explored; object, scene and action recognition with particular emphasis on designing highly effective and efficient feature representation and recognition algorithms.",
    keywords = "RGB-D object recognition, 3D recognition, Scene classification, Multi-modal network, Ego-centric action recognition",
    author = "{Mohd Zaki}, {Hasan Firdaus Bin}",
    year = "2017",
    doi = "10.4225/23/5a126d4cdf3fe",
    language = "English",
    school = "The University of Western Australia",

    }

    Mohd Zaki, HFB 2017, 'Object, scene and ego-centric action classification for robotic vision', Doctor of Philosophy, The University of Western Australia. https://doi.org/10.4225/23/5a126d4cdf3fe

    Object, scene and ego-centric action classification for robotic vision. / Mohd Zaki, Hasan Firdaus Bin.

    2017.

    Research output: ThesisDoctoral Thesis

    TY - THES

    T1 - Object, scene and ego-centric action classification for robotic vision

    AU - Mohd Zaki, Hasan Firdaus Bin

    PY - 2017

    Y1 - 2017

    N2 - Social robotics research focuses on developing intelligent machines that can perform difficult tasks especially those beyond human capabilities, involving repetitions or adverse conditions. However, robotic vision for high-level recognition associated with reasoning of uncertainty in unstructured environments is still far-fetched compared to the visual comprehension of humans. This dissertation investigates techniques to exploit the rich information from multi-modality sensors in pursuit of extending the frontiers of robotic vision. Three robot-centric recognition tasks are specifically explored; object, scene and action recognition with particular emphasis on designing highly effective and efficient feature representation and recognition algorithms.

    AB - Social robotics research focuses on developing intelligent machines that can perform difficult tasks especially those beyond human capabilities, involving repetitions or adverse conditions. However, robotic vision for high-level recognition associated with reasoning of uncertainty in unstructured environments is still far-fetched compared to the visual comprehension of humans. This dissertation investigates techniques to exploit the rich information from multi-modality sensors in pursuit of extending the frontiers of robotic vision. Three robot-centric recognition tasks are specifically explored; object, scene and action recognition with particular emphasis on designing highly effective and efficient feature representation and recognition algorithms.

    KW - RGB-D object recognition

    KW - 3D recognition

    KW - Scene classification

    KW - Multi-modal network

    KW - Ego-centric action recognition

    U2 - 10.4225/23/5a126d4cdf3fe

    DO - 10.4225/23/5a126d4cdf3fe

    M3 - Doctoral Thesis

    ER -