3D Scene Understanding at Urban Intersection using Stereo Vision and Digital Map

TL;DR

This paper presents a novel approach combining stereo vision and 3D digital maps to analyze urban traffic scenes for autonomous vehicles, enhancing obstacle detection, classification, and localization at intersections.

Contribution

It introduces a probabilistic framework integrating geometric, semantic, dynamic, and contextual cues for comprehensive urban intersection understanding.

Findings

Effective obstacle detection and tracking in urban scenes

Improved ego-localization using 3D digital maps

Validated on real traffic data from Tokyo

Abstract

The driving behavior at urban intersections is very complex. It is thus crucial for autonomous vehicles to comprehensively understand challenging urban traffic scenes in order to navigate intersections and prevent accidents. In this paper, we introduce a stereo vision and 3D digital map based approach to spatially and temporally analyze the traffic situation at urban intersections. Stereo vision is used to detect, classify and track obstacles, while a 3D digital map is used to improve ego-localization and provide context in terms of road-layout information. A probabilistic approach that temporally integrates these geometric, semantic, dynamic and contextual cues is presented. We qualitatively and quantitatively evaluate our proposed technique on real traffic data collected at an urban canyon in Tokyo to demonstrate the efficacy of the system in providing comprehensive awareness of the traffic surroundings.