Reliable reconstruction of passive non-line-of-sight imaging with occluder by deep learning

Yoosun Kim; Mooseok Jang

doi:10.1117/12.3017933

26 February 2024 Reliable reconstruction of passive non-line-of-sight imaging with occluder by deep learning

Yoosun Kim, Mooseok Jang

Proceedings Volume 13076, SPIE Advanced Biophotonics Conference (SPIE ABC 2023); 1307608 (2024) https://doi.org/10.1117/12.3017933
Event: SPIE Advanced Biophotonics Conference (SPIE ABC 2023), 2023, Jeju, Korea, Republic of

Abstract

Non-Line of Sight (NLoS) imaging represents a crucial technology with significant potential across numerous applications including autonomous vehicles, robotics, biomedicine, and state-of-the-art weaponry. NLoS technology primarily aims to rapidly and accurately reconstruct hidden objects following light perception. Existing methodologies predominantly employ active laser illumination. However, there is a burgeoning necessity for comprehensive research within passive NLoS systems, where objects self-emit light. In our study, we introduce a novel technique using a straightforward imaging setup composed of a standard digital camera and an occluder. By leveraging the power of deep learning, we demonstrate the capability to reconstruct hidden objects swiftly, significantly surpassing the time efficiency of existing techniques. Moreover, our method exhibits a high degree of reliability. Our research opens the door to further investigations into a universal and highly reliable passive NLoS reconstruction algorithm. This algorithm's defining feature is its resilience to changes in both the occluder and object positions within the imaging setup. Such imaging techniques, which eliminate the requirement for an auxiliary light source, offer promising prospects for the proactive detection of concealed adversaries or objects, thereby providing a valuable tool across various domains.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Yoosun Kim and Mooseok Jang "Reliable reconstruction of passive non-line-of-sight imaging with occluder by deep learning", Proc. SPIE 13076, SPIE Advanced Biophotonics Conference (SPIE ABC 2023), 1307608 (26 February 2024); https://doi.org/10.1117/12.3017933

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
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image restoration

Deep learning

Light sources

Non line of sight propagation

Systems modeling

Cameras

Reflection

Show All Keywords

Keywords/Phrases

Search In:

Publication Years