A study on depth map generation using a light field camera and a monocular RGB camera based on deep learning

Makoto Takamatsu; Makoto Hasegawa

doi:10.1117/12.2542653

16 October 2019 A study on depth map generation using a light field camera and a monocular RGB camera based on deep learning

Makoto Takamatsu, Makoto Hasegawa

Author Affiliations +

Proceedings Volume 11205, Seventh International Conference on Optical and Photonic Engineering (icOPEN 2019); 112050T (2019) https://doi.org/10.1117/12.2542653
Event: Seventh International Conference on Optical and Photonic Engineering (icOPEN 2019), 2019, Phuket, Thailand

Abstract

A depth map and an RGB image taken by a light field camera for training data are arranged in a dataset pair; the datasets are learnt through a deep learning method called pix2pix, which is a type of conditional generative adversarial network. We can generate depth maps using only a monocular mobile camera without the light field camera based on our proposed method. Low accuracy on depth is a technical issue for the light field camera; however, the proposed method improves the depth accuracy due to the generalization ability of neural networks.

Citation Download Citation

Makoto Takamatsu and Makoto Hasegawa "A study on depth map generation using a light field camera and a monocular RGB camera based on deep learning", Proc. SPIE 11205, Seventh International Conference on Optical and Photonic Engineering (icOPEN 2019), 112050T (16 October 2019); https://doi.org/10.1117/12.2542653

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