Last month, we had the opportunity to present our latest work in the field of machinelearning for computer vision during the ePictureThis workshop2023! This workshop was a great opportunity to show the current capability of the Research & Development team with Boris Lenseigne, Julia Cohen and Edouard Villain.
By aiming to improve human-machine interaction, this research focuses on the prediction of the user’s gaze using consumer-grade components. To do so, a wide range of artificial intelligence tools are used: from machine learning, to temporal neural networks.
The biggest challenge faced so far is the generalisation of the model for multiple users. However, current work about new architectures, and advanced preprocessing are tools that might be able to face it.
The end goal of this research project is to enable the use of computers for paraplegic users.
If you are interested in the whole presentation, you can find it here: https://www.researchgate.net/publication/372555560_Machine_learning_for_computer_vision_a_case_study_in_man_machine_interaction