Tri-Clustered Tensor Completion for Social-Aware Image Tag Refinement
Image Tag Refinement
Social image tag refinement, which aims to improve tag quality by automatically completing the missing tags and rectifying the noise-corrupted ones, is an essential component for social image search. Conventional approaches mainly focus on exploring the visual and tag information, without considering the user information, which often reveals important hints on the (in)correct tags of social images. Towards this end, we propose a novel tri-clustered tensor completion framework to collaboratively explore these three kinds of information to improve the performance of social image tag refinement. Specifically, the inter-relations among users, images and tags are modeled by a tensor, and the interrelations between users, images and tags are explored by three regularization respectively. To address the challenges of the super-sparse and large-scale tensor factorization that demands expensive computing and memory cost, tri-clustering method to divide the tensor into a certain number of sub-tensors by simultaneously clustering users, images and tags into a bunch of tri-clusters. And then we investigate two strategies to complete these sub-tensors by considering (in)dependence between the sub-tensors. Experimental results on a real-world social image database demonstrate the superiority of the proposed method compared with the state-of-the-art methods. On social websites, users are allowed to upload personal images, label them with freely-chosen tags, and join user groups with common interests. Due to the various professional backgrounds of users, their provided tags tend to be ambiguous, noisy and incomplete. If directly leveraging these noisy and incomplete social tags to perform the tag-based image retrieval, the performance will be far from satisfactory. Therefore, researchers are motivated to develop social image tag refinement approaches to improve the quality of social tags so as to reduce the semantic gap. This task is closely related to tag completion image (re)tagging and image annotation. The goal of social image tag refinement is to automatically complete the missing tags and rectify the noise-corrupted ones.
Get IEEE 2018 Project Topics List