DMCA
KONTSCHIEDER et al.: DISCRIMINATIVE CONTOUR FRAGMENT LEARNING 1 Discriminative Learning of Contour Fragments for Object Detection
BibTeX
@MISC{Kontschieder_kontschiederet,
author = {Peter Kontschieder and Hayko Riemenschneider and Michael Donoser and Horst Bischof},
title = {KONTSCHIEDER et al.: DISCRIMINATIVE CONTOUR FRAGMENT LEARNING 1 Discriminative Learning of Contour Fragments for Object Detection},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The goal of this work is to discriminatively learn contour fragment descriptors for the task of object detection. Unlike previous methods that incorporate learning techniques only for object model generation or for verification after detection, we present a holistic object detection system using solely shape as underlying cue. In the learning phase, we interrelate local shape descriptions (fragments) of the object contour with the corresponding spatial location of the object centroid. We introduce a novel shape fragment descriptor that abstracts spatially connected edge points into a matrix consisting of angular relations between the points. Our proposed descriptor fulfills important properties like distinctiveness, robustness and insensitivity to clutter. During detection, we hypothesize object locations in a generalized Hough voting scheme. The back-projected votes from the fragments allow to approximately delineate the object contour. We evaluate our method e.g. on the well-known ETHZ shape data base, where we achieve an average detection score of 87.5 % at 1.0 FPPI only from Hough voting, outperforming the highest scoring Hough voting approaches by almost 8%. 1
Keyphrases
object detection discriminative learning discriminative contour fragment learning contour fragment object contour object centroid corresponding spatial location scoring hough voting approach underlying cue holistic object detection system edge point well-known ethz shape data base object location angular relation average detection score object model generation generalized hough voting scheme matrix consisting novel shape fragment descriptor contour fragment descriptor important property back-projected vote local shape description hough voting previous method learning phase
