Learning pre-attentive driving behaviour from holistic visual features

Authors:
Nicolas Pugeault;Richard Bowden
Affiliations:
Centre for Vision, Speech and Signal Processing, University of Surrey, UK;Centre for Vision, Speech and Signal Processing, University of Surrey, UK
Venue:
ECCV'10 Proceedings of the 11th European conference on Computer vision: Part VI
Year:
2010

Citing 8
Cited 0

A decision-theoretic generalization of on-line learning and an application to boosting

Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Modeling the Shape of the Scene: A Holistic Representation of the Spatial Envelope

International Journal of Computer Vision
Contextual Priming for Object Detection

International Journal of Computer Vision
Robust Real-Time Face Detection

International Journal of Computer Vision
Rapid Biologically-Inspired Scene Classification Using Features Shared with Visual Attention

IEEE Transactions on Pattern Analysis and Machine Intelligence
Evaluation of GIST descriptors for web-scale image search

Proceedings of the ACM International Conference on Image and Video Retrieval
Biologically inspired mobile robot vision localization

IEEE Transactions on Robotics
Robot steering with spectral image information

IEEE Transactions on Robotics

Quantified Score

Hi-index	0.01

Visualization

Abstract

The aim of this paper is to learn driving behaviour by associating the actions recorded from a human driver with pre-attentive visual input, implemented using holistic image features (GIST). All images are labelled according to a number of driving-relevant contextual classes (eg, road type, junction) and the driver's actions (eg, braking, accelerating, steering) are recorded. The association between visual context and the driving data is learnt by Boosting decision stumps, that serve as input dimension selectors. Moreover, we propose a novel formulation of GIST features that lead to an improved performance for action prediction. The areas of the visual scenes that contribute to activation or inhibition of the predictors is shown by drawing activation maps for all learnt actions. We show good performance not only for detecting driving-relevant contextual labels, but also for predicting the driver's actions. The classifier's false positives and the associated activation maps can be used to focus attention and further learning on the uncommon and difficult situations.