Department of Electrical Engineering (ISY)

Combining Shadow Detection and Simulation for Estimation of Vehicle Size and Position

B. Johansson, J. Wiklund, P.-E. Forssén, G. Granlund
Pattern Recognition Letters
Volume 30, Number 8, Pages 751-759
June 2009

Abstract

This paper presents a method that combines shadow detection and a 3D box model including shadow simulation, for estimation of size and position of vehicles. We define a similarity measure between a simulated image of a 3D box, including the box shadow, and a captured image that is classified into background/foreground/shadow. The similarity measure is used in an optimization procedure to find the optimal box state. It is shown in a number of experiments and examples how the combination shadow detection/simulation improves the estimation compared to just using detection or simulation, especially when the shadow detection or the simulation is inaccurate. We also describe a tracking system that utilizes the estimated 3D boxes, including highlight detection, spatial window instead of a time based window for predicting heading, and refined box size estimates by weighting accumulated estimates depending on view. Finally we show example results.

Keywords

Vehicle tracking, 3D box model, object size estimation, shadow detection, shadow simulation

Full Paper

This paper is electronically available from ScienceDirect

---

Bibtex entry

@Article{johansson09,
  author = 	 {B. Johansson and J. Wiklund and P.-E. Forss{\'e}n and G. Granlund},
  title = 	 {Combining Shadow Detection and Simulation for Estimation of Vehicle Size and Position},
  journal = 	 {Pattern Recognition Letters},
  year = 	 {2009},
  volume = 	 {30},
  number = 	 {8},
  pages = 	 {751-759},
  month = 	 {June},
  issn =         {0167-8655},
  publisher =    {Elsevier Ltd.},
  note =         {On line March 20 2009.}
}