Image annotation by k NN-sparse graph-based label propagation over noisily tagged web images

• DOI: 10.1145/1899412.1899418
• Publication Date: 2012-10-12T20:56:02Z
• Authors: Jinhui Tang; Richang Hong; Shuicheng Yan; Tat-Seng Chua; Guo-Jun Qi; Ramesh Jain
• Keywords: Discriminative model; Regularization; Dense graph

In this article, we exploit the problem of annotating a large-scale image corpus by label propagation over noisily tagged web images. To annotate images more accurately, propose novel k NN-sparse graph-based semi-supervised learning approach for harnessing labeled and unlabeled data simultaneously. The sparse graph constructed datum-wise one-vs- NN reconstructions all samples can remove most semantically unrelated links among data, thus it is robust discriminative than conventional graphs. Meanwhile, apply approximate nearest neighbors to accelerate construction without loosing its effectiveness. More importantly, an effective training refinement strategy within framework handle noise in labels, bringing dual regularization both quantity sparsity noise. We conduct extensive experiments on real-world database consisting 55,615 Flickr labels. results demonstrate effectiveness efficiency proposed capability deal with