Widening siamese architectures for stereo matching

被引：19

作者：

Brandao, Patrick ^{[1
,2
]}

Mazomenos, Evangelos ^{[1
,2
]}

Stoyanov, Danail ^{[1
]}

机构：

[1] UCL, Wellcome EPSRC Ctr Intervent & Surg Sci WEISS, Charles Bell House,Foley St, London W1W 7TS, England

[2] UCL, Dept Comp Sci, Charles Bell House,Foley St, London W1W 7TS, England

来源：

PATTERN RECOGNITION LETTERS | 2019年 / 120卷

基金：

英国工程与自然科学研究理事会; 欧盟地平线“2020”;

关键词：

Stereo matching; Convolutional neural network; Disparity; Computer vision;

D O I：

10.1016/j.patrec.2018.12.002

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Computational stereo is one of the classical problems in computer vision. Numerous algorithms and solutions have been reported in recent years focusing on developing methods for computing similarity, aggregating it to obtain spatial support and finally optimizing an energy function to find the final disparity. In this paper, we focus on the feature extraction component of stereo matching architecture and we show standard CNNs operation can be used to improve the quality of the features used to find point correspondences. Furthermore, we use a simple space aggregation that hugely simplifies the correlation learning problem, allowing us to better evaluate the quality of the features extracted. Our results on benchmark data are compelling and show promising potential even without refining the solution. (C) 2018 The Authors. Published by Elsevier B.V.

引用

页码：75 / 81

页数：7

共 17 条

[1]

Abadi M, 2016, PROCEEDINGS OF OSDI'16: 12TH USENIX SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION, P265

[2] Advances in computational stereo [J].

Brown, MZ ;

Burschka, D ;

Hager, GD .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2003, 25 (08) :993-1008

[3]

Fischer P., 2015, ICCV

[4]

Geiger A, 2012, PROC CVPR IEEE, P3354, DOI 10.1109/CVPR.2012.6248074

[5] End-to-End Learning of Geometry and Context for Deep Stereo Regression [J].

Kendall, Alex ;

Martirosyan, Hayk ;

Dasgupta, Saumitro ;

Henry, Peter ;

Kennedy, Ryan ;

Bachrach, Abraham ;

Bry, Adam .

2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2017, :66-75

[6]

Kingma DP, 2014, ADV NEUR IN, V27

[7] End-to-End Training of Hybrid CNN-CRF Models for Stereo [J].

Knoebelreiter, Patrick ;

Reinbacher, Christian ;

Shekhovtsov, Alexander ;

Pock, Thomas .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :1456-1465

[8]

LUO WJ, 2016, PROC CVPR IEEE, P5695, DOI DOI 10.1109/CVPR.2016.614

[9]

Menze M, 2015, PROC CVPR IEEE, P3061, DOI 10.1109/CVPR.2015.7298925

[10]

Pang J., 2017, ICCVW

← 1 2 →