supported by: Regione Toscana
Our smart audio guide perceives the context and is able to interact with users: it performs automatic recognition of artworks, to enable a semi-automatic interaction with the wearer. The system is backed by a computer vision system capable to work in real-time on a mobile device, coupled with audio and motion sensors. The system has been deployed on a NVIDIA Jetson TK1 and a NVIDIA Shield Tablet K1, and tested in a real world environment (Bargello Museum of Florence).
Video available here.
Funded by: Fondazione Cassa di Risparmio Firenze
The main project’s aim is to study and develop a methodology to identify mobile devices through the analysis and combination of signals coming from the various on-board sensors, aiming to extract a smartphone fingerprint which is univocal and distinctive of one specific device.
We developed a comprehensive treatise of the three task of image tag assignment, refinement, and tag-based image retrieval. We introduce the functionality of tag relevance, i.e. estimating the relevance of a tag with respect to the visual content of a given image and its social context. A two-dimensional taxonomy to structure the growing literature and understand the ingredients of the main works is developed. A new experimental protocol is presented and eleven representative works are implemented and evaluated. Putting all this together, this work aims to provide an overview of the past and foster progress for the near future.