Title
Saliency Detection in a Virtual Driving Environment for Autonomous Vehicle Behavior Improvement
Abstract
To make the best decisions in real-world situations, autonomous vehicles require learning algorithms that process a large number of labeled images. This paper aims to compare the automatically generated saliency maps with attention maps obtained with an eye-tracking device in order to provide automated labeling of images for the learning algorithm. To simulate traffic scenarios, we are using a virtual driving environment with a motion platform and an eye-tracking device for identifying the driver's attention. The saliency maps are generated by post-processing the driver's view provided by the front camera.
Year
DOI
Venue
2021
10.1007/978-3-030-87595-4_37
AUGMENTED REALITY, VIRTUAL REALITY, AND COMPUTER GRAPHICS
Keywords
DocType
Volume
Driving simulator, Autonomous vehicle, Saliency map
Conference
12980
ISSN
Citations 
PageRank 
0302-9743
0
0.34
References 
Authors
0
5