Weakly supervised visual dictionary learning by harnessing image attributes

CORC > 厦门大学 > 信息技术－已发表论文

	Weakly supervised visual dictionary learning by harnessing image attributes
	Gao, Yue ; Ji, Rongrong ; Liu, Wei ; Dai, Qionghai ; Hua, Gang ; Ji RR(纪荣嵘)
刊名	http://dx.doi.org/10.1109/TIP.2014.2364536
	2014
关键词	Computer vision Hidden Markov models Image retrieval Image segmentation Markov processes Semantics Supervised learning
英文摘要	Bag-of-features (BoFs) representation has been extensively applied to deal with various computer vision applications. To extract discriminative and descriptive BoF, one important step is to learn a good dictionary to minimize the quantization loss between local features and codewords. While most existing visual dictionary learning approaches are engaged with unsupervised feature quantization, the latest trend has turned to supervised learning by harnessing the semantic labels of images or regions. However, such labels are typically too expensive to acquire, which restricts the scalability of supervised dictionary learning approaches. In this paper, we propose to leverage image attributes to weakly supervise the dictionary learning procedure without requiring any actual labels. As a key contribution, our approach establishes a generative hidden Markov random field (HMRF), which models the quantized codewords as the observed states and the image attributes as the hidden states, respectively. Dictionary learning is then performed by supervised grouping the observed states, where the supervised information is stemmed from the hidden states of the HMRF. In such a way, the proposed dictionary learning approach incorporates the image attributes to learn a semantic-preserving BoF representation without any genuine supervision. Experiments in large-scale image retrieval and classification tasks corroborate that our approach significantly outperforms the state-of-the-art unsupervised dictionary learning approaches.
语种	英语
出版者	Institute of Electrical and Electronics Engineers Inc.
内容类型	期刊论文
源URL	[http://dspace.xmu.edu.cn/handle/2288/92940]
专题	信息技术－已发表论文
推荐引用方式 GB/T 7714	Gao, Yue,Ji, Rongrong,Liu, Wei,et al. Weakly supervised visual dictionary learning by harnessing image attributes[J]. http://dx.doi.org/10.1109/TIP.2014.2364536,2014.
APA	Gao, Yue,Ji, Rongrong,Liu, Wei,Dai, Qionghai,Hua, Gang,&纪荣嵘.(2014).Weakly supervised visual dictionary learning by harnessing image attributes.http://dx.doi.org/10.1109/TIP.2014.2364536.
MLA	Gao, Yue,et al."Weakly supervised visual dictionary learning by harnessing image attributes".http://dx.doi.org/10.1109/TIP.2014.2364536 (2014).