Automatic Photo Annotation Based on Visual Content Analysis

In this paper we describe a novel approach for automatic photo annotation based on visual content analysis. We combine local and global features to ensure robustness and generalization needed by complex queries. We place great emphasis to work in real-time. To cope with the huge number of extracted features, we implemented diskbased locality-sensitive hashing to index descriptors. By searching candidates to extraction of keywords, we focus on photos analysis in terms of probability, that the retrieved photos contain the right keywords for the target photo. The result is that we are able to name key objects directly in the target photo.