Please use this identifier to cite or link to this item:
https://doi.org/10.1109/ICME.2010.5583896
Title: | Evaluation of histogram based interest point detector in web image classification and search | Authors: | Cai, J. Zha, Z.-J. Zhao, Y. Wang, Z. |
Keywords: | Bag of visual word Histogram based Interest point detector Local feature |
Issue Date: | 2010 | Citation: | Cai, J., Zha, Z.-J., Zhao, Y., Wang, Z. (2010). Evaluation of histogram based interest point detector in web image classification and search. 2010 IEEE International Conference on Multimedia and Expo, ICME 2010 : 613-618. ScholarBank@NUS Repository. https://doi.org/10.1109/ICME.2010.5583896 | Abstract: | Local image feature has received increasing attention in various applications, such as web image classification and search. The process of local feature extraction consists of two main steps: interest point detection and local feature description. A wealth of interest point detectors have been proposed in last decades. Most of them measure pixel-wise differences in image intensity or color. Recently, a new type of interest point detector has been developed, which incorporates histogram-based representation into the process of interest point detection. In this paper, we evaluate this histogram-based interest point detector in the context of web image classification and search, as well as compare it against typical pixel-based detectors and heuristic grid-based detector. The evaluation is performed on two web image datasets: NUS-WIDE-OBJECT and MIRFLICKR-25000 datasets. The experimental results demonstrate that the histogram-based interest point detector outperforms the pixelbased and grid-based detectors in both web image classification and search tasks. © 2010 IEEE. | Source Title: | 2010 IEEE International Conference on Multimedia and Expo, ICME 2010 | URI: | http://scholarbank.nus.edu.sg/handle/10635/40453 | ISBN: | 9781424474912 | DOI: | 10.1109/ICME.2010.5583896 |
Appears in Collections: | Staff Publications |
Show full item record
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.