Concepedia

Learning-based Optical Flow

Learning-Based Optical Flow is an optical flow estimation method that belongs to the class of data-driven approaches and is distinguished by its reliance on machine learning models, particularly deep neural networks, to learn the mapping from image pairs to optical flow fields through training on large datasets.

174

Publications

12.9K

Citations

649

Authors

248

Institutions

Publications per year

2017–2026

141

Parents

Publications

174

PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume

Deqing Sun, Xiaodong Yang, Ming-Yu Liu et al. · 2018 · 2.7K citations

, , +14

DVC: An End-To-End Deep Video Compression Framework

Guo Lu, Wanli Ouyang, Dong Xu et al. · 2019 · 703 citations

, , +16

Optical flow modeling and computation: A survey

Denis Fortun, Patrick Bouthémy, Charles Kervrann · Computer Vision and Image Understanding · 2015 · 413 citations · Full text

, , +11

GMFlow: Learning Optical Flow via Global Matching

Haofei Xu, Jing Zhang, Jianfei Cai et al. · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · 2022 · 363 citations · Full text

, , +12

On Benchmarking Optical Flow

Brendan McCane, Kevin Novins, D Crannitch et al. · Computer Vision and Image Understanding · 2001 · 252 citations

, , +10

1–5 of 174

Authors

649

  1. DS
    Deqing Sun

    Google (United States), United States

    4 publications · 2.9K citations

    4

    h-index

  2. GD
    Guido de Croon

    Delft University of Technology, Netherlands

    4 publications · 264 citations

    4

    h-index

  3. WS
    Wolfgang Schröder

    RWTH Aachen University, Germany

    3 publications · 181 citations

    3

    h-index

  4. KL
    Kai Lagemann

    University of Cambridge, United Kingdom

    3 publications · 181 citations

    3

    h-index

  5. SL
    Shuaicheng Liu

    University of Electronic Science and Technology of China, China

    3 publications · 104 citations

    3

    h-index

1–5 of 649

Institutions

248

  1. Chinese University of Hong Kong

    Hong Kong, Hong Kong

    5 h-index

    15 publications

    978 citations

  2. Shanghai Jiao Tong University

    Shanghai, China

    5 h-index

    18 publications

    3.1K citations

  3. University College London

    London, United Kingdom

    5 h-index

    11 publications

    397 citations

  4. Hong Kong University of Science and Technology

    Hong Kong, Hong Kong

    5 h-index

    12 publications

    392 citations

  5. University of Cambridge

    Cambridge, United Kingdom

    5 h-index

    6 publications

    271 citations

  6. Google (United States)

    Mountain View, United States

    4 h-index

    15 publications

    1.2K citations

  7. Carnegie Mellon University

    Pittsburgh, United States

    4 h-index

    5 publications

    132 citations

  8. Zhejiang University

    Hangzhou, China

    4 h-index

    12 publications

    303 citations

1–8 of 248

Venues

  1. arXiv (Cornell University)

    Repository · 15 publications · 400 citations

    11

    h-index

  2. Computer Vision and Image Understanding

    CVIU · 7 publications · 934 citations

    7

    h-index

  3. Sensors

    5 publications · 170 citations

    5

    h-index

  4. 5

    h-index

  5. IEEE Transactions on Image Processing

    4 publications · 334 citations

    4

    h-index

1–5