DMCA
Feature-centric efficient subwindow search (2009)
Cached
Download Links
| Venue: | In ICCV |
| Citations: | 16 - 5 self |
BibTeX
@INPROCEEDINGS{Lehmann09feature-centricefficient,
author = {Alain Lehmann and Bastian Leibe and Luc Van Gool and Computer Vision Laboratory},
title = {Feature-centric efficient subwindow search},
booktitle = {In ICCV},
year = {2009}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Many object detection systems rely on linear classifiers embedded in a sliding-window scheme. Such exhaustive search involves massive computation. Efficient Subwindow Search (ESS) [11] avoids this by means of branch and bound. However, ESS makes an unfavourable memory tradeoff. Memory usage scales with both image size and overall object model size. This risks becoming prohibitive in a multiclass system. In this paper, we make the connection between slidingwindow and Hough-based object detection explicit. Then, we show that the feature-centric view of the latter also nicely fits with the branch and bound paradigm, while it avoids the ESS memory tradeoff. Moreover, on-line integral image calculations are not needed. Both theoretical and quantitative comparisons with the ESS bound are provided, showing that none of this comes at the expense of performance. 1.
Keyphrases
feature-centric efficient subwindow search image size es memory tradeoff linear classifier feature-centric view exhaustive search memory usage scale es bound many object detection system overall object model size hough-based object detection explicit on-line integral image calculation quantitative comparison sliding-window scheme massive computation efficient subwindow search unfavourable memory tradeoff multiclass system bound paradigm
