Zhou, J., Bischof, W. F. and Caelli, T. (2005). Robust and efficient road tracking in aerial images. Proceedings of the Workshop on Object Extraction for 3D City Models, Road Databases and Traffic Monitoring - Concepts, Algorithms, and Evaluation,Vienna, Austria, August 29-30.

Automated road tracking is important for map revision but is currently not reliable enough to be useful for industrial applications. Consequently semi-automatic road tracking has become the preferred solution. In this paper we introduce a road tracking system based on particle filtering and human-computer interactions. Particle filters were used to estimate road axis points. During the estimation and human-computer interaction, new reference profiles were generated and stored in the road template memory for future correlation analysis, thus covering the space of road profiles. Human input provided the road tracker with initial estimates, updated state parameters and multiple reference profiles. This approach has resulted in remarkable improvements in efficiency, compared to the human-only approach while preserving robustness and accuracy.

Back to publications.