We present VEDI (Vision Exploitation for Data Interpretation), an integrated system to jointly assist the visitors of cultural sites and provide meaningful statistics about the cultural site to managers. To address both goals, VEDI includes a wearable assistant (implemented through a wearable device such as HoloLens) which leverages Computer Vision algorithms to understand where the user is and what they are paying attention to. At the visitor's end, such information is leveraged to augment the visit by displaying additional information on the observed points of interest, helping the visitors to navigate the site and suggesting what to see next. Concurrently, a back-end extracts high-level behavioral information from the captured video content, which is used to provide the site manager with meaningful statistics and performance indexes on the cultural site. Experiments show that VEDI achieves good results on both the indoor and outdoor cultural sites considered for the experimentation.
VEDI architecture is composed of 4 main parts: 1) Mobile devices, 2) GPU, 3) Charging and update station, 4) Central system.
G.M. Farinella, G. Signorello, S. Battiato, A. Furnari, F. Ragusa, R. Leonardi, E. Ragusa, E. Scuderi, A. Lopes, L. Santo, M. Samarotto. VEDI: Vision Exploitation for Data Interpretation. In 20th International Conference on Image Analysis and Processing, 2019. Download the paper here.