Using the shadow as a single feature for real-time monocular vehicle pose determination

Authors:
Dennis Rosebrock;Markus Rilk;Jens Spehr;Friedrich M. Wahl
Affiliations:
Institut für Robotik und Prozessinformatik, Technische Universität Braunschweig, Braunschweig, Germany;Institut für Robotik und Prozessinformatik, Technische Universität Braunschweig, Braunschweig, Germany;Institut für Robotik und Prozessinformatik, Technische Universität Braunschweig, Braunschweig, Germany;Institut für Robotik und Prozessinformatik, Technische Universität Braunschweig, Braunschweig, Germany
Venue:
ISVC'11 Proceedings of the 7th international conference on Advances in visual computing - Volume Part I
Year:
2011

Citing 5
Cited 1

Intensity and Edge-Based Symmetry Detection Applied to Car-Following

ECCV '92 Proceedings of the Second European Conference on Computer Vision
Detecting Moving Shadows: Algorithms and Evaluation

IEEE Transactions on Pattern Analysis and Machine Intelligence
On-Road Vehicle Detection: A Review

IEEE Transactions on Pattern Analysis and Machine Intelligence
Combining shadow detection and simulation for estimation of vehicle size and position

Pattern Recognition Letters
Introduction to Algorithms, Third Edition

Introduction to Algorithms, Third Edition

A feature construction method for general object recognition

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this work we propose a way to detect vehicles in monocular camera images and determine their position and orientation on the ground plane relative to the camera. The camera does not need to be stationary which allows the method to be used in mobile applications. Its results can therefore serve as an input to advanced driver assistance systems (ADAS). The single feature used is the shadow beneath the vehicles. We implemented a real-time applicable method to detect these shadows under strongly varying conditions and determine the corresponding vehicle pose. Finally we evaluate our results by comparing them to ground truth data.