SSN15 / Predicting-visual-search-targets-from-eye-movements

Academic project on Machine Learning, Deep Learning and Computer vision

Motivation:

1. Is it possible to infer the persons goals by analyzing their eye fixations?
2. Eye movements can reveal complex cognitive states of mind
3. Analyse Visual Behavior
4. Closed World Vs Open World Settings
5. Prediction of Search Targets using Stationary Eye Tracker

Hardware Mobile eye tracker

Data

• Data collected from ~18 participants
• Virtual objects

Data Preprocessing

• Implemented computer vision algorithms to preprocess data

Feature extraction

• Used pre-trained network like resNet-50 to extract features from data

Approaches

• Implemented various Machine learning/Deep Learning algorithms to predict the search target

For more information on the project, please go through the slides in the following link: https://github.com/SSN15/Predicting-visual-search-targets-from-eye-movements/blob/master/HCI_05Feb2019.pdf