Kawanabe, M.M.KawanabeBinder, A.A.BinderMüller, C.C.MüllerWojcikiewicz, W.W.Wojcikiewicz2022-03-112022-03-112011https://publica.fraunhofer.de/handle/publica/37376910.1109/WACV.2011.5711531Automatic annotation of images is a challenging task in computer vision because of "semantic gap" between high-level visual concepts and image appearances. Therefore, user tags attached to images can provide further information to bridge the gap, even though they are partially un-informative and misleading. In this work, we investigate multi-modal visual concept classification based on visual features and user tags via kernel-based classifiers. An issue here is how to construct kernels between sets of tags. We deploy Markov random walks on graphs of key tags to incorporate co-occurrence between them. This procedure acts as a smoothing of tag based features. Our experimental result on the ImageCLEF2010 PhotoAnnotation benchmark shows that our proposed method outperforms the baseline relying solely on visual information and a recently published state-of-the-art approach.en004conference paper