As usual, I am sorry Mauro for devastating your day by finding research projects that overlap or already did what you want to do in your phd but this is really interesting: Leveraging context for mobile media metadata by Marc Davis , Nathan Good, Risto Sarvas.

The recent popularity of mobile camera phones allows for new opportunities to gather important metadata at the point of capture. This paper describes and demonstrates a method for generating metadata for images using spatial, temporal, and social context. We describe a system we implemented for inferring location information for pictures taken with camera phones. We propose that leveraging contextual metadata at the point of capture can bridge the problems of the semantic and sensory gaps. In particular, combining and sharing spatial, temporal, and social contextual metadata from a given user and across users allows us to make inferences about media content .