A system to recommend dishes by the real time recognition of dining activity

Enhancement of dining environment by information technology is explored in this paper. A table system to recommend additional dishes according to the dining status is introduced. By modeling dining behavior as the sequence of eating, dining status is estimated from the history of bites. The system recognizes the user's dining activity in real time, and stores its history. It is expected to be useful in the environment where additional dishes are often ordered such as a pub. By iterative design, the hand recognition method in the system was changed from the use of a visual marker in the first prototype to image processing in the current prototype. Evaluation was conducted to show the improvement of the recognition.