Share Email Print
cover

Proceedings Paper

Building robust appearance models using on-line feature selection
Author(s): R. Porter; R. Loveland; E. Rosten
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In many tracking applications, adapting the target appearance model over time can improve performance. This approach is most popular in high frame rate video applications where latent variables, related to the objects appearance (e.g., orientation and pose), vary slowly from one frame to the next. In these cases the appearance model and the tracking system are tightly integrated, and latent variables are often included as part of the tracking system's dynamic model. In this paper we describe our efforts to track cars in low frame rate data (1 frame / second), acquired from a highly unstable airborne platform. Due to the low frame rate, and poor image quality, the appearance of a particular vehicle varies greatly from one frame to the next. This leads us to a different problem: how can we build the best appearance model from all instances of a vehicle we have seen so far. The best appearance model should maximize the future performance of the tracking system, and maximize the chances of reacquiring the vehicle once it leaves the field of view. We propose an online feature selection approach to this problem and investigate the performance and computational trade-offs with a real-world dataset.

Paper Details

Date Published: 9 April 2007
PDF: 9 pages
Proc. SPIE 6574, Optical Pattern Recognition XVIII, 657409 (9 April 2007); doi: 10.1117/12.721439
Show Author Affiliations
R. Porter, Los Alamos National Lab. (United States)
R. Loveland, Los Alamos National Lab. (United States)
E. Rosten, Los Alamos National Lab. (United States)


Published in SPIE Proceedings Vol. 6574:
Optical Pattern Recognition XVIII
David P. Casasent; Tien-Hsin Chao, Editor(s)

© SPIE. Terms of Use
Back to Top