Compressive Non-line-of-sight Imaging using a Convolutional Neural Network

Shenyu Zhu; Yong Meng Sua; Ting Bu; Yu Ping Huang

Compressive Non-line-of-sight Imaging using a Convolutional Neural Network

Shenyu Zhu
, Yong Meng Sua
, Ting Bu
, Yu Ping Huang

Department of Physics

Stevens Institute of Technology

Research output: Contribution to journal › Conference article › peer-review

Abstract

We demonstrate compressive non-line-of-sight imaging with downsampling ratio of 6.25% by using a convolutional neural network (CNN). Photon arrival-time histogram with 10 picosecond resolution enables high-quality image reconstruction with CNN trained purely by using simulated dataset.

Original language	English
Article number	JW4A.69
Journal	Optics InfoBase Conference Papers
State	Published - 2022
Event	Frontiers in Optics, FiO 2022 - Rochester, United States Duration: 17 Oct 2022 → 20 Oct 2022

Cite this

@article{5fffd17ab72245bbbeb6d867d4bbc8d9,

title = "Compressive Non-line-of-sight Imaging using a Convolutional Neural Network",

abstract = "We demonstrate compressive non-line-of-sight imaging with downsampling ratio of 6.25\% by using a convolutional neural network (CNN). Photon arrival-time histogram with 10 picosecond resolution enables high-quality image reconstruction with CNN trained purely by using simulated dataset.",

author = "Shenyu Zhu and Sua, \{Yong Meng\} and Ting Bu and Huang, \{Yu Ping\}",

note = "Publisher Copyright: {\textcopyright} 2022 The Author(s); Frontiers in Optics, FiO 2022 ; Conference date: 17-10-2022 Through 20-10-2022",

year = "2022",

language = "English",

}

TY - JOUR

T1 - Compressive Non-line-of-sight Imaging using a Convolutional Neural Network

AU - Zhu, Shenyu

AU - Sua, Yong Meng

AU - Bu, Ting

AU - Huang, Yu Ping

PY - 2022

Y1 - 2022

N2 - We demonstrate compressive non-line-of-sight imaging with downsampling ratio of 6.25% by using a convolutional neural network (CNN). Photon arrival-time histogram with 10 picosecond resolution enables high-quality image reconstruction with CNN trained purely by using simulated dataset.

AB - We demonstrate compressive non-line-of-sight imaging with downsampling ratio of 6.25% by using a convolutional neural network (CNN). Photon arrival-time histogram with 10 picosecond resolution enables high-quality image reconstruction with CNN trained purely by using simulated dataset.

UR - https://www.scopus.com/pages/publications/85146800764

UR - https://www.scopus.com/pages/publications/85146800764#tab=citedBy

M3 - Conference article

AN - SCOPUS:85146800764

JO - Optics InfoBase Conference Papers

JF - Optics InfoBase Conference Papers

M1 - JW4A.69

T2 - Frontiers in Optics, FiO 2022

Y2 - 17 October 2022 through 20 October 2022

ER -

Compressive Non-line-of-sight Imaging using a Convolutional Neural Network

Abstract

Other files and links

Fingerprint

Cite this