COSTA: Co-Occurrence Statistics for Zero-Shot Classification

Thomas Mensink, Efstratios Gavves, Cees G.M. Snoek; The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014, pp. 2441-2448

Abstract


In this paper we aim for zero-shot classification, that is visual recognition of an unseen class by using knowledge transfer from known classes. Our main contribution is COSTA, which exploits co-occurrences of visual concepts in images for knowledge transfer. These inter-dependencies arise naturally between concepts, and are easy to obtain from existing annotations or web-search hit counts. We estimate a classifier for a new label, as a weighted combination of related classes, using the co-occurrences to define the weight. We propose various metrics to leverage these co-occurrences, and a regression model for learning a weight for each related class. We also show that our zero-shot classifiers can serve as priors for few-shot learning. Experiments on three multi-labeled datasets reveal that our proposed zero-shot methods, are approaching and occasionally outperforming fully supervised SVMs. We conclude that co-occurrence statistics suffice for zero-shot classification.

Related Material


[pdf]
[bibtex]
@InProceedings{Mensink_2014_CVPR,
author = {Mensink, Thomas and Gavves, Efstratios and Snoek, Cees G.M.},
title = {COSTA: Co-Occurrence Statistics for Zero-Shot Classification},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2014}
}