Non-line-of-sight imaging based on low-memory global convolutional neural network

Yaojia Yang; Enlai Guo; Jing Han

doi:10.1117/12.3057156

22 January 2025 Non-line-of-sight imaging based on low-memory global convolutional neural network

Yaojia Yang, Enlai Guo, Jing Han

Author Affiliations +

Proceedings Volume 13520, 7th Optics Young Scientist Summit (OYSS 2024); 135200Q (2025) https://doi.org/10.1117/12.3057156
Event: Seventh Optics Young Scientist Summit (OYSS 2024), 2024, Nanjing, China

Abstract

Non-Line-of-Sight (NLOS) imaging aims to reconstruct targets or scenes that are directly invisible. When it comes to the research of time-of-flight-based transient NLOS techniques, due to the 3D nature of transient measurements, existing methods require high computational memory, which is not feasible. In this paper, we provide a lightweight U-Net architecture that performs well with low computational cost requirements when reconstructing a hidden target object. Our method makes use of complex convolutional layers with gate mechanisms and channel attention, enabling efficient extraction of local feature details and global information. Extensive experiments show that our method outperforms current methods on simulated data and achieves comparable performance to current methods on real-world data.

(2025) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Yaojia Yang, Enlai Guo, and Jing Han "Non-line-of-sight imaging based on low-memory global convolutional neural network", Proc. SPIE 13520, 7th Optics Young Scientist Summit (OYSS 2024), 135200Q (22 January 2025); https://doi.org/10.1117/12.3057156

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