Non-line-of-sight-imaging using dynamic relay surfaces

Abstract

The non-line-of-sight (NLOS) imaging problem has attracted a lot of interest in recent years. The objective is to produce images of objects that are hidden around a corner, using the information encoded in the time-of-flight (ToF) of photons that scatter multiple times after incidence at a given relay surface. Most current methods assume a Lambertian, flat and static relay surface, with non-moving targets in the hidden scene. Here we show NLOS reconstructions for a relay surface that is non-planar and rapidly changing during data acquisition. Our NLOS imaging system exploits two different detectors to collect the ToF data; one pertaining to the relay surface and another one regarding the ToF information of the hidden scene. The system is then able to associate where the multiply-scattered photons originated from the relay surface. This step allows us to account for changing relay positions in the reconstruction algorithm. Results show that the reconstructions for a dynamic relay surface are similar to the ones obtained using a traditional non-dynamic relay surface.

Document Details

Document Type
Pub Defense Publication
Publication Date
Feb 12, 2020
Source ID
10.1364/oe.383586

Entities

People

  • Andreas Velten
  • Ji-hyun Nam
  • Marco La Manna
  • Syed Azer Reza

Organizations

  • Defense Advanced Research Projects Agency
  • National Aeronautics and Space Administration
  • Office of Naval Research

Tags

Fields of Study

  • Physics

Readers

  • Electromagnetic Wave Scattering and Antenna Radiation Engineering
  • Neural Network Machine Learning.
  • Sensor Fusion and Tracking Systems.