DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch

    Abstract

    Our goal is to significantly speed up the runtime of current state-of-the-art stereo algorithms to enable real-time inference. Towards this goal, we developed a differentiable PatchMatch module that allows us to discard most disparities without requiring full cost volume evaluation. We then exploit this representation to learn which range to prune for each pixel. By progressively reducing the search space and effectively propagating such information, we are able to efficiently compute the cost volume for high likelihood hypotheses and achieve savings in both memory and computation. Finally, an image guided refinement module is exploited to further improve the performance. Since all our components are differentiable, the full network can be trained end-to-end. Our experiments show that our method achieves competitive results on KITTI and SceneFlow datasets while running in real-time at 62ms.

    Authors

    Shivam Duggal, Shenlong Wang, Wei-Chiu Ma, Rui Hu, Raquel Urtasun

    Conference

    ICCV 2019

    Full Paper

    ‘DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch’ (PDF)

    Uber ATG

    Comments
    Previous articleMaximum Relevance and Minimum Redundancy Feature Selection Methods for a Marketing Machine Learning Platform
    Next articleDeconstructing Lottery Tickets: Zeros, Signs, and the Supermask
    Shivam Duggal
    Shivam Duggal is an AI Resident at Uber ATG Toronto supervised by Prof. Raquel Urtasun. Before joining ATG, he received his bachelors degree in computer science from Delhi Technological University in 2017 and then worked as an engineer in Amazon focusing on Machine Learning, alongside doing independent research. His research interests are Computer Vision and Deep Learning.
    Shenlong Wang
    Shenlong Wang is research scientist at Uber ATG Toronto working on the development of self-driving cars. He is also a PhD student at University of Toronto. His advisor is Prof. Raquel Urtasun. He has a broad interest in computer vision, machine learning and robotics. He is particularly interested in 3D vision and deep structured models.
    Wei-Chiu Ma
    Wei-Chiu Ma is a PhD student at MIT advised by Prof. Antonio Torralba. His research interests lie in the intersection of computer vision and machine learning, in particular low-level vision and 3D vision. He also works part-time at Uber ATG Toronto with Prof. Raquel Urtasun to apply his research to self-driving vehicles.
    Rui Hu
    Rui Hu is a research engineer at Uber ATG Toronto. Before that he was a staff engineer at Qualcomm and senior engineer at Apple SPG. His research interests include neural networks, computer vision and GPU optimization.
    Raquel Urtasun
    Raquel Urtasun is the Chief Scientist for Uber ATG and the Head of Uber ATG Toronto. She is also a Professor at the University of Toronto, a Canada Research Chair in Machine Learning and Computer Vision and a co-founder of the Vector Institute for AI. She is a recipient of an NSERC EWR Steacie Award, an NVIDIA Pioneers of AI Award, a Ministry of Education and Innovation Early Researcher Award, three Google Faculty Research Awards, an Amazon Faculty Research Award, a Connaught New Researcher Award, a Fallona Family Research Award and two Best Paper Runner up Prize awarded CVPR in 2013 and 2017. She was also named Chatelaine 2018 Woman of the year, and 2018 Toronto’s top influencers by Adweek magazine