Tags tagging

Social media sharing webs allow consumers to describe the media content with tags. However, tagging all images in a personal album in detail is a time-consuming task and assigning unified tags to the whole album simply will greatly degrade the tagging accuracy. In this paper, a novel scheme to automatically tagging personal albums is proposed. For a personal album, an affinity propagation algorithm is first adopted to obtain a set of representative images. Then, both visual information and semantic information are exploited to the estimate the relevance scores of tags for representative images, and a random walk algorithm is adopted to refine the obtained relevance scores. Finally, tags of the rest images in the personal album are automatically obtained based on a graph-based semi-supervise learning method. In such way, a good trade-off between the number of tagged photos and high tagging accuracy can be achieved. The experimental results demonstrate the effectiveness of the proposed scheme.