@inproceedings {pub2606,
	title = {A Two-stage Classifier Architecture for Detecting Objects under Real-world Occlusion Patterns},
	author = {Marvin Struwe AND Stephan Hasler AND Ute Bauer-Wersing},
	year = {2014},
	abstract = {Despite extensive efforts, state-of-the-art detection approaches show a strong
degradation of performance with increasing level of occlusion. In this paper we
investigate a strategy to improve the detection of occluded objects based
on the analytic feature framework from \cite{struwe13} and compare the
results in a car detection task. Motivated by an analysis of annotated traffic
scenes we focus on a general concept to handle vertical occlusion patterns. For
this we describe a two stage classifier architecture that detects
vertical car parts in the first stage and combines the local responses in the
second. As an extension we provide depth information for the individual car
parts helping the classifier in the second stage to reason about typical
occlusion patterns. The results reveal that this general strategy can provide a
substantial performance improvement for appearance-based detection approaches.},
	publisher = {Marvin Struwe, Stephan Hasler, and Ute Bauer-Wersing},
	booktitle = {International Conference on Artificial Neural Networks ICANN 2014 }
}