GPU-accelerated real-time stixel computation

TL;DR

This paper presents a GPU-accelerated implementation of the Stixel World scene representation, enabling real-time processing on embedded devices and high-end GPUs with high accuracy and efficiency.

Contribution

It introduces a complete GPU-based pipeline for stixel estimation, achieving real-time performance on embedded hardware and high throughput on powerful GPUs.

Findings

Real-time stixel estimation at 26 fps on Tegra X1

Over 400 fps on Titan X GPU

Reliable scene representation at high resolution

Abstract

The Stixel World is a medium-level, compact representation of road scenes that abstracts millions of disparity pixels into hundreds or thousands of stixels. The goal of this work is to implement and evaluate a complete multi-stixel estimation pipeline on an embedded, energy-efficient, GPU-accelerated device. This work presents a full GPU-accelerated implementation of stixel estimation that produces reliable results at 26 frames per second (real-time) on the Tegra X1 for disparity images of 1024x440 pixels and stixel widths of 5 pixels, and achieves more than 400 frames per second on a high-end Titan X GPU card.