Researchers at HSE University have taken an important step in the field of eye movement data analysis by creating a set of algorithms that is now available to everyone. This development will greatly simplify the work of scientists and developers of artificial intelligence systems that use eye trackers — devices that record where a person is looking while performing various tasks. This was reported by the press service of HSE University.
Our library can be useful to researchers, as it allows not only to repeat what was already available in other software, but also to apply new algorithms and create more powerful models for research.
In recent years, eye trackers have become popular among psychologists and sociologists who study how people react to various stimuli. These devices require special software that tracks the movement of the pupils and records which objects a person's attention is focused on.
The algorithms developed by Russian scientists work in several stages. First, they analyze the video and determine when the pupils are fixated on certain objects. Then, the program evaluates how long the eyes remain on these objects, which helps to predict further eye movements.
According to Surkov, the development may be particularly useful in marketing research, allowing to assess how consumers react to advertising and what elements attract their attention.
These algorithms were created as part of the strategic project "AI Technologies for Humans" ("Priority-2030"). Thanks to the open source code, they can be easily integrated into any research or commercial projects using eye trackers or similar technologies.
Read also:
Ceramic shield to protect patients' eyes during radiation therapy created in Russia
Volgograd scientists develop AI-module LensCalc to improve cataract replacement surgeries
