A Discriminative Kernel-Based Approach to Rank Images from Text Queries

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 30; no. 8; pp. 1371 - 1384
• Main Authors: Grangier, D., Bengio, S.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01-08-2008; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Artificial Intelligence; Construction; Discriminant Analysis; discriminative learning; Distance learning; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Image retrieval; Information Storage and Retrieval - methods; Kernel; kernel-based classifier; Labeling; large margin; Learning; Natural Language Processing; Pattern Recognition, Automated - methods; Performance evaluation; Photography; Predictive models; Queries; Ranking; Retrieval; Search engines; Studies; Tasks; Texts; Vocabulary; Vocabulary, Controlled; Web search; Wire; image retrieval; kernel-based classifier; large margin; ranking; discriminative learning
• ISSN: 0162-8828; 1939-3539
• DOI: 10.1109/TPAMI.2007.70791

This paper introduces a discriminative model for the retrieval of images from text queries. Our approach formalizes the retrieval task as a ranking problem, and introduces a learning procedure optimizing a criterion related to the ranking performance. The proposed model hence addresses the retrieval problem directly and does not rely on an intermediate image annotation task, which contrasts with previous research. Moreover, our learning procedure builds upon recent work on the online learning of kernel-based classifiers. This yields an efficient, scalable algorithm, which can benefit from recent kernels developed for image comparison. The experiments performed over stock photography data show the advantage of our discriminative ranking approach over state-of-the-art alternatives (e.g. our model yields 26.3% average precision over the Corel dataset, which should be compared to 22.0%, for the best alternative model evaluated). Further analysis of the results shows that our model is especially advantageous over difficult queries such as queries with few relevant pictures or multiple-word queries.