Int20h hackathon results

The team of students of the MMDA department successfully performed at the Int20h hackathon and took 4th place! In the Data Science section, they managed to develop the perfect solution in 20 hours: splitting images into unlabeled clusters for learning, and generating averaged faces for the resulting clusters that reflect gender, ethnic groups, emotions, and even stylistic features such as glasses or makeup. But the most interesting thing is that they were able to transfer the image into two-dimensional space, which allowed to display the unique characteristics of different groups. Our students have shown that there are no impossible tasks, only new challenges to achieve great results!