Please use this identifier to cite or link to this item: https://doi.org/10.1109/ICASSP.2007.366073
Title: Outlier detection from pooled data for image retrieval system evaluation
Authors: Xiong, W.
Ong, S.H. 
Lim, J.H.
Tian, Q.
Xu, C.
Zhang, N.
Foong, K. 
Keywords: Image classification
Image recognition
Pattern classification
Pattern clustering
Pattern recognition
Issue Date: 2007
Source: Xiong, W.,Ong, S.H.,Lim, J.H.,Tian, Q.,Xu, C.,Zhang, N.,Foong, K. (2007). Outlier detection from pooled data for image retrieval system evaluation. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings 1 : I977-I980. ScholarBank@NUS Repository. https://doi.org/10.1109/ICASSP.2007.366073
Abstract: Widely used in the evaluation of retrieval systems, the pooling method collects top ranked images from submitted retrieval, systems resulting in possibly a very large pool Of images. Inevitably, the pool may contain outliers, Human experts then manually annotate the relevance of them to create a ground truth for evaluation. Studies show that this annotation is time-consuming, tedious and inconsistent. To reduce human workload, this paper introduces an automatic method to detect outliers. Different from traditional detection methods using unsupervised techniques only, we utilize both supervised and unsupervised techniques sequentially as both positive and negative examples are (partially) available in this context. Specifically, support vector machines (SVMs) and fuzzy c-means clustering are used to predict data relevance and "outlier-ness". Performance improvements using our method after outlier removal have been validated on the medical image retrieval taskin ImageCLEF 2004. © 2007 IEEE.
Source Title: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
URI: http://scholarbank.nus.edu.sg/handle/10635/71347
ISBN: 1424407281
ISSN: 15206149
DOI: 10.1109/ICASSP.2007.366073
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Page view(s)

40
checked on Dec 9, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.