Skip to Main content Skip to Navigation
Conference papers

Learning to Rank Using High-Order Information

Abstract : The problem of ranking a set of visual samples according to their relevance to a query plays an important role in computer vision. The traditional approach for ranking is to train a binary classifier such as a support vector machine (svm). Binary classifiers suffer from two main deficiencies: (i) they do not optimize a ranking-based loss function, for example, the average precision (ap) loss; and (ii) they cannot incorporate high-order information such as the a priori correlation between the rele-vance of two visual samples (for example, two persons in the same image tend to perform the same action). We propose two novel learning formu-lations that allow us to incorporate high-order information for ranking. The first framework, called high-order binary svm (hob-svm), allows for a structured input. The parameters of hob-svm are learned by minimiz-ing a convex upper bound on a surrogate 0-1 loss function. In order to obtain the ranking of the samples that form the structured input, hob-svm sorts the samples according to their max-marginals. The second framework, called high-order average precision svm (hoap-svm), also allows for a structured input and uses the same ranking criterion. How-ever, in contrast to hob-svm, the parameters of hoap-svm are learned by minimizing a difference-of-convex upper bound on the ap loss. Using a standard, publicly available dataset for the challenging problem of action classification, we show that both hob-svm and hoap-svm outperform the baselines that ignore high-order information.
Complete list of metadata

Cited literature [25 references]  Display  Hide  Download
Contributor : Puneet Kumar Dokania Connect in order to contact the contributor
Submitted on : Tuesday, October 21, 2014 - 3:04:58 PM
Last modification on : Friday, January 21, 2022 - 3:01:31 AM
Long-term archiving on: : Thursday, January 22, 2015 - 10:36:08 AM


Files produced by the author(s)




Puneet Kumar Dokania, A. Behl, C. V. Jawahar, M Pawan Kumar. Learning to Rank Using High-Order Information. ECCV 2014 - European Conference on Computer Vision (2014), Sep 2014, Zurch, Switzerland. pp.609 - 623, ⟨10.1007/978-3-319-10593-2_40⟩. ⟨hal-01076220⟩



Record views


Files downloads