Robust Optical Flow in Rainy Scenes

被引:26
|
作者
Li, Ruoteng [1 ]
Tan, Robby T. [1 ,2 ]
Cheong, Loong-Fah [1 ]
机构
[1] Natl Univ Singapore, Singapore, Singapore
[2] Yale NUS Coll, Singapore, Singapore
来源
COMPUTER VISION - ECCV 2018, PT 15 | 2018年 / 11219卷
关键词
Optical flow; Rain; Decomposition; Residue channel;
D O I
10.1007/978-3-030-01267-0_18
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Optical flow estimation in rainy scenes is challenging due to degradation caused by rain streaks and rain accumulation, where the latter refers to the poor visibility of remote scenes due to intense rainfall. To resolve the problem, we introduce a residue channel, a single channel (gray) image that is free from rain, and its colored version, a colored-residue image. We propose to utilize these two rain-free images in computing optical flow. To deal with the loss of contrast and the attendant sensitivity to noise, we decompose each of the input images into a piecewise-smooth structure layer and a high-frequency fine-detail texture layer. We combine the colored-residue images and structure layers in a unified objective function, so that the estimation of optical flow can be more robust. Results on both synthetic and real images show that our algorithm outperforms existing methods on different types of rain sequences. To our knowledge, this is the first optical flow method specifically dealing with rain. We also provide an optical flow dataset consisting of both synthetic and real rain images.
引用
收藏
页码:299 / 317
页数:19
相关论文
共 50 条
  • [41] Evaluation of Clustering Methods for Finding Dominant Optical Flow Fields in Crowded Scenes
    Eibl, Guenther
    Braendle, Norbert
    19TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOLS 1-6, 2008, : 2531 - 2534
  • [42] Robust optical flow estimation using invariant feature
    Zhou, Q
    Zeng, YJ
    Chelberg, D
    Starzyk, J
    PROCEEDINGS OF THE 33RD SOUTHEASTERN SYMPOSIUM ON SYSTEM THEORY, 2001, : 263 - 267
  • [43] Robust Interpolation of Correspondences for Large Displacement Optical Flow
    Hu, Yinlin
    Li, Yunsong
    Song, Rui
    30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, : 4791 - 4799
  • [44] Graph Laplacian Regularization for Robust Optical Flow Estimation
    Young, Sean, I
    Naman, Aous T.
    Taubman, David
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2020, 29 : 3970 - 3983
  • [45] Dense optical flow via robust data fusion
    Kharbat, Mohd
    Aouf, Nabil
    SIGNAL IMAGE AND VIDEO PROCESSING, 2011, 5 (02) : 203 - 215
  • [46] ANISOTROPIC FLUID SOLVER FOR ROBUST OPTICAL FLOW SMOOTHING
    Doshi, Ashish
    Bors, Adrian G.
    2009 10TH INTERNATIONAL WORKSHOP ON IMAGE ANALYSIS FOR MULTIMEDIA INTERACTIVE SERVICES, 2009, : 117 - 120
  • [47] Robust Bioinspired Architecture for Optical-Flow Computation
    Botella, Guillermo
    Garcia, Antonio
    Rodriguez-Alvarez, Manuel
    Ros, Eduardo
    Meyer-Baese, Uwe
    Molina, Maria C.
    IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2010, 18 (04) : 616 - 629
  • [48] Blur robust optical flow using motion channel
    Li, Wenbin
    Chen, Yang
    Lee, JeeHang
    Ren, Gang
    Cosker, Darren
    NEUROCOMPUTING, 2017, 220 : 170 - 180
  • [49] Two-stage robust optical flow estimation
    Ye, M
    Haralick, RM
    IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, VOL II, 2000, : 623 - 628
  • [50] Uncertainty optimization for robust dynamic optical flow estimation
    Willert, Volker
    Toussaint, Marc
    Eggert, Julian
    Koerner, Edgar
    ICMLA 2007: SIXTH INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND APPLICATIONS, PROCEEDINGS, 2007, : 450 - +