DMCA
Occlusion Patterns for Object Class Detection
Cached
Download Links
| Citations: | 13 - 1 self |
BibTeX
@MISC{Pepik_occlusionpatterns,
author = {Bojan Pepik and Michael Stark and Peter Gehler Bernt Schiele},
title = {Occlusion Patterns for Object Class Detection},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Despite the success of recent object class recognition systems, the long-standing problem of partial occlusion re-mains a major challenge, and a principled solution is yet to be found. In this paper we leave the beaten path of meth-ods that treat occlusion as just another source of noise – instead, we include the occluder itself into the modelling, by mining distinctive, reoccurring occlusion patterns from annotated training data. These patterns are then used as training data for dedicated detectors of varying sophistica-tion. In particular, we evaluate and compare models that range from standard object class detectors to hierarchical, part-based representations of occluder/occludee pairs. In an extensive evaluation we derive insights that can aid fur-ther developments in tackling the occlusion challenge. 1.
Keyphrases
occlusion pattern object class detection standard object class detector training data principled solution annotated training data occluder occludee pair occlusion challenge dedicated detector long-standing problem partial occlusion fur-ther development beaten path extensive evaluation recent object class recognition system part-based representation major challenge
