Development of Machine Learning Models for IR-Based Eye Tracking - 2026_IDR_DEIB_19
Scientific-Disciplinary Group
09/IINF-03 - Telecommunications
Description
The research aims to develop an advanced eye-tracking system based on Photo-Reflective Sensor Oculography (PSOG), addressing calibration and cross-user generalization challenges through the use of hypernetworks. PSOG signals are highly sensitive to geometric, anatomical, and illumination variations, which introduce non-stationary shifts and degrade the performance of conventional models. The project proposes the use of hypernetworks to dynamically adapt the parameters of the gaze estimator based on calibration embeddings, reducing calibration complexity. Validation will be conducted on multi-source PSOG datasets with controlled perturbations and diverse operating conditions. The final objective is to achieve a more stable, adaptive, and generalizable PSOG system with fast calibration and robust performance under domain variations.
Job posting website
Number of positions
1
Funding body
Politecnico di Milano
Selection process
Click to expand
View the original posting on the MUR website: Go to MUR website