Robust Eye-Gaze Estimation Deep Neural Network (REGENN)

• Unité de recherche: INNOSUISSE
• Numéro de projet: 26041.1 PFES-ES
• Titre du projet anglais: Robust Eye-Gaze Estimation Deep Neural Network (REGENN)

Description succincte (Allemand)

Robust Eye-Gaze Estimation Deep Neural Network (REGENN)

Description succincte (Anglais)

Robust Eye-Gaze Estimation Deep Neural Network (REGENN)

Résumé des résultats (Abstract) (Allemand)

Eyeware creates natural and intuitive human-machine interaction experiences using eye-gaze tracking. We propose to design a novel, robust gaze estimation system based on multi-task deep learning that can adapt to challenging conditions, targeting new applications in home automation and advertising.

Résumé des résultats (Abstract) (Anglais)

Eyeware creates natural and intuitive human-machine interaction experiences using eye-gaze tracking. We propose to design a novel, robust gaze estimation system based on multi-task deep learning that can adapt to challenging conditions, targeting new applications in home automation and advertising.