A 3D reconstruction algorithm based on adaptive cost aggregation and depth map optimization

Yulin Wang; Tao Song; Xiaoling Xu; Gang Mou; Zhengdong Li

doi:10.1117/12.2656738

20 October 2022 A 3D reconstruction algorithm based on adaptive cost aggregation and depth map optimization

Yulin Wang, Tao Song, Xiaoling Xu, Gang Mou, Zhengdong Li

Proceedings Volume 12451, 5th International Conference on Computer Information Science and Application Technology (CISAT 2022); 1245144 (2022) https://doi.org/10.1117/12.2656738
Event: 5th International Conference on Computer Information Science and Application Technology (CISAT 2022), 2022, Chongqing, China

Abstract

Aiming at the problems of large memory consumption, long calculation time and difficult reconstruction in weak texture and occlusion areas of the existing deep learning 3D reconstruction network, this paper proposes a cascade 3D reconstruction network algorithm based on phased estimation of resolution. First, a visibility-aware adaptive cost aggregation method is introduced to aggregate the cost volume, and the visibility of pixels in the view is obtained through the network; Secondly, the disparity range per pixel is predicted based on variance to construct a spatially varying depth hypothesis surface; Finally, an improved depth map fusion algorithm is used to check the consistency of the re-projection error between the pixel point and the 3D point, and a dense point cloud is obtained. The quantitative and qualitative comparison with other methods on the DTU data set shows that the algorithm in this paper can reconstruct scenes with better performance in detail and reduce GPU memory consumption and computing time.

Citation Download Citation

Yulin Wang, Tao Song, Xiaoling Xu, Gang Mou, and Zhengdong Li "A 3D reconstruction algorithm based on adaptive cost aggregation and depth map optimization", Proc. SPIE 12451, 5th International Conference on Computer Information Science and Application Technology (CISAT 2022), 1245144 (20 October 2022); https://doi.org/10.1117/12.2656738

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

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
10 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Reconstruction algorithms

3D modeling

Computer vision technology

Machine vision

Optimization (mathematics)

3D image processing

Pattern recognition

Show All Keywords

Keywords/Phrases

Search In:

Publication Years