Single-Shot Real-Time Multiple-Path Time-of-Flight Depth Imaging for Multi-Aperture and Macro-Pixel Sensors
Miguel Heredia Conde, Keiichiro Kagawa, Tomoya Kokado, Shoji Kawahito, Otmar Loffeld
Abstract
Multiple-Path Interference (MPI) is a major drawback of Time-of-Flight (ToF) sensors. Current methods resolving more than a single light bounce per pixel rely on the sequential acquisition of large amounts of data and are computationally expensive. These factors have precluded the development of a real-time multiple-path ToF camera to date. In this work we consider two hardware alternatives able to acquire all necessary raw data in a single shot. We introduce a unified signal processing pipeline that takes the single-shot raw data from any of these sensors as input and delivers an independent depth image per interfering path. We use the measurements to estimate the lowest-frequency Fourier coefficients of the scene response function, which can be ideally modeled as a train of Dirac delta functions. A robust parametric spectral estimation method allows us recovering the unknown parameters (distance and reflectivity of the targets) for all pixels in real time.
BibTeX
@inproceedings{icassp2020_singleshotrealti,
title = {Single-Shot Real-Time Multiple-Path Time-of-Flight Depth Imaging for Multi-Aperture and Macro-Pixel Sensors},
author = {Miguel Heredia Conde and Keiichiro Kagawa and Tomoya Kokado and Shoji Kawahito and Otmar Loffeld},
booktitle = {ICASSP 2020},
year = {2020}
}