Multiple candidates and multiple constraints based accurate depth estimation for multi-view stereo

Chao Zhang; Fugen Zhou; Bindang Xue

doi:10.1117/12.2267110

8 February 2017 Multiple candidates and multiple constraints based accurate depth estimation for multi-view stereo

Chao Zhang, Fugen Zhou, Bindang Xue

Proceedings Volume 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016); 102251O (2017) https://doi.org/10.1117/12.2267110
Event: Eighth International Conference on Graphic and Image Processing, 2016, Tokyo, Japan

Abstract

In this paper, we propose a depth estimation method for multi-view image sequence. To enhance the accuracy of dense matching and reduce the inaccurate matching which is produced by inaccurate feature description, we select multiple matching points to build candidate matching sets. Then we compute an optimal depth from a candidate matching set which satisfies multiple constraints (epipolar constraint, similarity constraint and depth consistency constraint). To further increase the accuracy of depth estimation, depth consistency constraint of neighbor pixels is used to filter the inaccurate matching. On this basis, in order to get more complete depth map, depth diffusion is performed by neighbor pixels’ depth consistency constraint. Through experiments on the benchmark datasets for multiple view stereo, we demonstrate the superiority of proposed method over the state-of-the-art method in terms of accuracy.

Citation Download Citation

Chao Zhang, Fugen Zhou, and Bindang Xue "Multiple candidates and multiple constraints based accurate depth estimation for multi-view stereo", Proc. SPIE 10225, Eighth International Conference on Graphic and Image Processing (ICGIP 2016), 102251O (8 February 2017); https://doi.org/10.1117/12.2267110

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available