Non-line-of-sight imaging using phasor-field virtual wave optics

Abstract

This supplemental document contains the following information:A Overview of the method B Derivation of the phasor field C LOS template functions D Implementation details of the RSD solvers

Phasor field diffraction based reconstruction for fast non-line-of-sight imaging systems

“…The conventional RSD propagates the electric field, but in this context propagation of an intensity modulation is required. The phasor field RSD differs from the conventional version by the amplitude correction factor α(x v ) 10 . This factor depends on the location x v of the reconstruction point and could be precomputed once the geometry of the relay surface is known.…”

“…Liu et al 10 and Reza et al 11 introduce a virtual wave phasor field formalism that is the basis of this work. Using the phasor field method, the NLOS imaging problem can be stated as a line of sight optical imaging problem based on diffraction and solved using existing diffraction theory methods.…”

Non-Line-of-Sight Imaging

“…Over the last few years, a variety of different approaches to addressing the NLOS problem have been proposed. Some of these focus on advanced measurement systems, using femtosecond and picosecond time-resolved detectors [2][3][4][5], interferometry [6,7], acoustic systems [8], passive imaging systems [9][10][11], or thermal imaging system [12,13], while others explore various models of light transport that make certain assumptions on reflectance or other properties of the hidden scenes. At the convergence of physics, signal processing, optics, and electronics, NLOS imaging is an interdisciplinary challenge that has seen much progress over the last few years.…”

“…It still seems unclear, however, what the "best" representation for general NLOS imaging is. Wave Optics Models, rather than the above outlined geometric optics model, have recently been explored for transient imaging configurations with time-resolved detectors and pulsed light sources [4,5,[86][87][88][89] (see Figs. 4,5).…”

“…The entire scan involves 24000 points leading to data collection times of 4 minutes, 2 minutes, and 24 seconds respectively. The scene is about 2 meters wide and 3 meters deep.Figure adapted from[4]…”