Zero-shot image categorization by image correlation exploration

Gao, LianLi, Song, Jingkuan, Shao, Junming, Zhu, Xiaofeng and Shen, HengTao (2015). Zero-shot image categorization by image correlation exploration. In: ICMR’15: Proceedings of the 2015 ACM International Conference on Multimedia Retrieval. 5th ACM on International Conference on Multimedia Retrieval, Shanghai, China, (487-490). 23-26 June. doi:10.1145/2671188.2749309


Author Gao, LianLi
Song, Jingkuan
Shao, Junming
Zhu, Xiaofeng
Shen, HengTao
Title of paper Zero-shot image categorization by image correlation exploration
Conference name 5th ACM on International Conference on Multimedia Retrieval
Conference location Shanghai, China
Conference dates 23-26 June
Convener Hauptmann, Alex
Proceedings title ICMR’15: Proceedings of the 2015 ACM International Conference on Multimedia Retrieval
Journal name ICMR 2015 - Proceedings of the 2015 ACM International Conference on Multimedia Retrieval
Place of Publication New York, NY United States
Publisher Association for Computing Machinery
Publication Year 2015
Year available 2015
Sub-type Fully published paper
DOI 10.1145/2671188.2749309
Open Access Status Not Open Access
ISBN 9781450332743
Start page 487
End page 490
Total pages 4
Chapter number 68
Total chapters 114
Collection year 2016
Language eng
Abstract/Summary The problem of image categorization from zero or only a few training examples, called zero-shot learning, occurs frequently, but it has hardly been studied in computer vision research. To tackle this problem, mid-level semantic attributes are introduced to identify image categories. For example, one can construct a classifier for the giant panda category by enumerating its attributes (e.g., black, white and four-footed) even without providing giant panda training images. Recently, several studies have investigated to learn attribute classifiers, based on which new classes can be detected. However, an often-encountered problem is the limited number of training data due to the time-consuming manual annotation of the attributes. Also, using single feature is hard to detect some attributes, e.g., the HSV feature is not robust enough to predict 'tusk' or 'flies' attributes. In this paper, we propose a unified semi-supervised learning (SSL) framework that learns the attribute classifiers by utilizing multiple feature and exploring the correlations between images. Specifically, we learn an optimal graph which embeds the relationships among the data points more accurately. Then, this graph is used to generate a geometrical regularizers for a semi-supervised learning model to learn the attribute classifier by utilizing both labeled and unlabeled images. Afterward, new classes can be detected based on their attribute representation. The use of SSL can boost the performances of attribute classifiers with very few training examples, and the adoption of multiple features makes the attribute prediction more robust. Experimental results on a series of real benchmark data sets suggest that semi-supervised learning do enhance the performances of attribute prediction and zero-shot categorization, compared with state-of-the-art methods.
Q-Index Code E1
Q-Index Status Provisional Code
Institutional Status UQ
Additional Notes This conference is renamed from ACM SIGMM International Workshop on Multimedia Information Retrieval, which is ranked B.

 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in Thomson Reuters Web of Science Article
Scopus Citation Count Cited 2 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Mon, 25 Jan 2016, 20:47:33 EST by Dr Heng Tao Shen on behalf of School of Information Technol and Elec Engineering