DMCA
Integrated Pedestrian Classification and Orientation Estimation
Cached
Download Links
| Citations: | 19 - 4 self |
BibTeX
@MISC{Enzweiler_integratedpedestrian,
author = {Markus Enzweiler and Dariu M. Gavrila},
title = {Integrated Pedestrian Classification and Orientation Estimation},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
This paper presents a novel approach to single-frame pedestrian classification and orientation estimation. Unlike previous work which addressed classification and orientation separately with different models, our method involves a probabilistic framework to approach both in a unified fashion. We address both problems in terms of a set of view-related models which couple discriminative expert classifiers with sample-dependent priors, facilitating easy integration of other cues (e.g. motion, shape) in a Bayesian fashion. This mixture-of-experts formulation approximates the probability density of pedestrian orientation and scalesup to the use of multiple cameras. Experiments on large real-world data show a significant performance improvement in both pedestrian classification and orientation estimation of up to 50%, compared to stateof-the-art, using identical data and evaluation techniques. 1.
Keyphrases
orientation estimation pedestrian classification probabilistic framework bayesian fashion identical data single-frame pedestrian classification pedestrian orientation easy integration different model large real-world data previous work novel approach unified fashion multiple camera mixture-of-experts formulation discriminative expert classifier view-related model sample-dependent prior significant performance improvement probability density evaluation technique
