ImageFeatures

Publication at: http://www.clef-initiative.eu/documents/71612/cbfdcac6-ebc9-40da-9825-e53e38a594b3

This system presents the participation of our group in the ImageCLEF 2012 Photo Annotation Task. The main objective of this task is to address the problem of automated image annotation by asking the participants to create systems that can automatically assign an image a subset of concepts from a list of 94 possible visual concepts. As images can be annotated with more than one concept we experiment with different strategies in order to extract the semantics inside them. We do this by using not only local image feature descriptors (TopSurf), but also try analyzing the colors (Color Moments), textures and shapes (Profile Entropy Features) and even make use of the user defined tag semantics and face detection algorithms.