Project EyeFeatures: an open source Python library for analysing eye movements

Project leader: Anton Surkov

Project team: Vagiz Daudov, Vladimir Zakharov, Daniil Tikhonov, Elena Artemenko

The project team won the competition for young scientists of the National Research University Higher School of Economics and the strategic project AI-technologies for humans in 2024. The main goal of the project is to create a unique tool for analysing oculomotor data. The library capabilities include oculomotor data processing, extraction of both classical features (fixations, saccades and their characteristics) and complex metrics (various types of entropy and other indicators). The module for comparing eye movement trajectories using different metrics may be of particular interest to researchers. The library also provides functionality for working with zones of interest and panel data structure. The deep module allows the use of specialised deep methods. Eye movement data is finding more and more applications, so the library will be useful for research in medicine, cognitive science and marketing. The source code, documentation, as well as installation instructions and tutorials can be found here.

The library is registered in Rospatent. At the moment, for further development of the library, intern researchers and developers are being recruited. It is planned to write scientific articles for submission to high-ranked journals.