Project Abstract

Predicting goal motivated human gaze is important in Human Computer Interactions (HCI). With the release of Meta?��s Quest line of headsets and the recent release of Apple Vision Pro, the future of many applications will depend on interactions between humans and computers through eyes and hand movement which will be far from current touch based interactions. Projects like these will augment applications and devices into an era of hands-free interaction and usability. Another advantage will be the battery life and the cost of AR/VR headsets. Currently, all available headsets available in the market suffer from poor battery life. A large part of this can be attributed to the number of cameras and scanners that check upon eye movements of the wearer. With gaze prediction models, we may be able to reduce these numbers increasing the battery life of the headset. The headsets may also become cheaper with this. This project aims to come up with a machine learning model that can predict goal oriented human gaze. We will focus on LSTMs to create such a model. And compare our model to existing models.

Keywords: Machine Learning, Saliency, Human Computer Interaction

 Conference Details

Session: Presentation Stream 24 at Presentation Slot 4

Location: College 017 at Wednesday 8th 09:00 – 12:30

Markers: Fabio Caraffini, Nader Al Khatib (GTA)

Course: MSc Computer Science, Masters PG

Future Plans: I’m looking for work