Learning with deep neural networks forms the state-of-the-art in many tasks such as image classification, image detection, speech recognition, text analysis. We here set out to gain understanding in learning in an ‘end-to-end’ manner for an autonomous vehicle, which refers to directly learning the decision which will result from the perception of the scene. For example, we consider learning a binary ‘stop’/‘go’ decision, with respect to pedestrians, given the input image. In this work we propose to use additional information, referred to as ‘proxy supervision’, for improved learning and study its effects on the overall performance. We show that the proxy labels significantly improve the robustness of learning, while achieving as good, or better, accuracy than in the original task of binary classification.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Learning with proxy supervision for end-to-end visual learning


    Contributors:


    Publication date :

    2017-06-01


    Size :

    1387167 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Tutor-based learning of visual categories using different levels of supervision

    Fritz, M. / Kruijff, G. J. / Schiele, B. | British Library Online Contents | 2010


    Learning to Recognize Objects with Little Supervision

    Carbonetto, P. / Dorkó, G. / Schmid, C. et al. | British Library Online Contents | 2008


    Interactive multiple object learning with scanty human supervision

    Villamizar, Michael / Garrell, Anaís / Sanfeliu, Alberto et al. | British Library Online Contents | 2016


    Interactive multiple object learning with scanty human supervision

    Villamizar, Michael / Garrell, Anaís / Sanfeliu, Alberto et al. | British Library Online Contents | 2016


    Interactive multiple object learning with scanty human supervision

    Villamizar, Michael / Garrell, Anaís / Sanfeliu, Alberto et al. | British Library Online Contents | 2016