Intelligence and learning proposal

Title: Restaurant recommending application using facial expression detection


Clmtrackr is a javascript library for fitting facial models to faces in videos or images. Clmtrackr tracks a face and outputs the coordinate positions of the face model as an array, following the numbering of the mode. 

I will use CLMTRACKR to track users’ facial expressions and rank 9 photos of food from different restaurants for them. The user flow is like this:

– Users can type in up to 3 different parameters in hashtag search boxes (city/ name of food/ type of food).

– The app will show (recommend) the users 9 photos of food from different restaurants for 3 seconds each.

– The app will analyze the users’ facial expressions using CLMTRACKR.

– The app will rank the 9 photos of food based on the data. The more the users smile, the higher the rank is.

– Based on the rank the users can click each photo to see info page for each restaurant. It tells you name of the venue, location (map with address), and one line comment on the food or the restaurant.

– Or, the app automatically turns into info page if the users smile (if the smile level is more than 0.5).



Leave a Reply

Your email address will not be published. Required fields are marked *