End-to-End learning of cost-volume aggregation for real-time dense stereo

Andrey Kuzmin, Dmitry Mikushin, Victor Lempitsky

    Результат исследований: Глава в книге, отчете, сборнике статейМатериалы для конференциирецензирование

    16 Цитирования (Scopus)

    Аннотация

    We present a new deep learning-based approach for dense stereo matching. Compared to previous works, our approach does not use deep learning of pixel appearance descriptors, employing very fast classical matching scores instead. At the same time, our approach uses a deep convolutional network to predict the local parameters of cost volume aggregation process, which in this paper we implement using differentiable domain transform. By treating such transform as a recurrent neural network, we are able to train our whole system that includes cost volume computation, cost-volume aggregation (smoothing), and winner-takes-all disparity selection end-to-end. The resulting method is highly efficient at test time, while achieving good matching accuracy. On the KITTI 2012 and KITTI 2015 benchmark, it achieves a result of 5.08% and 6.34% error rate respectively while running at 29 frames per second rate on a modern GPU.

    Язык оригиналаАнглийский
    Название основной публикации2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017 - Proceedings
    РедакторыNaonori Ueda, Jen-Tzung Chien, Tomoko Matsui, Jan Larsen, Shinji Watanabe
    ИздательIEEE Computer Society
    Страницы1-6
    Число страниц6
    ISBN (электронное издание)9781509063413
    DOI
    СостояниеОпубликовано - 5 дек. 2017
    Событие2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017 - Tokyo, Япония
    Продолжительность: 25 сент. 201728 сент. 2017

    Серия публикаций

    НазваниеIEEE International Workshop on Machine Learning for Signal Processing, MLSP
    Том2017-September
    ISSN (печатное издание)2161-0363
    ISSN (электронное издание)2161-0371

    Конференция

    Конференция2017 IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2017
    Страна/TерриторияЯпония
    ГородTokyo
    Период25/09/1728/09/17

    Fingerprint

    Подробные сведения о темах исследования «End-to-End learning of cost-volume aggregation for real-time dense stereo». Вместе они формируют уникальный семантический отпечаток (fingerprint).

    Цитировать