Project in collaboration with Gestalt ReVision group. In this project, we want to understand how people enjoy art while in an art exhibition. To do, one of the main sources of information is recordings of a mobile eye tracker of each participant while enjoying the art exhibition. This specific of the project is to develop an efficient pipeline to extract meaningful data from these recordings using Computer Vision techniques. The main steps of the pipeline are:
• Image classification
• Segmentation of different behaviours fusing video, accelerometer and eye-tracking data
• Image registration to map eye tracking coordinates to reference coordinates