Learning Super-resolved Depth from Active Gated Imaging

TL;DR

This paper introduces a method to generate high-resolution depth maps for autonomous driving by learning a mapping from active gated imaging data, achieving 5% relative accuracy between 25-80 meters.

Contribution

It presents a novel approach to obtain super-resolved depth maps using cost-effective active gated imaging and a learned mapping, addressing resolution and accuracy trade-offs.

Findings

Achieves 5% relative depth accuracy between 25-80 meters.

Produces super-resolved depth maps aligned with pixel intensities.

Utilizes low-cost diode and CMOS technology for depth sensing.

Abstract

Environment perception for autonomous driving is doomed by the trade-off between range-accuracy and resolution: current sensors that deliver very precise depth information are usually restricted to low resolution because of technology or cost limitations. In this work, we exploit depth information from an active gated imaging system based on cost-sensitive diode and CMOS technology. Learning a mapping between pixel intensities of three gated slices and depth produces a super-resolved depth map image with respectable relative accuracy of 5% in between 25-80 m. By design, depth information is perfectly aligned with pixel intensity values.