The Computer vision team will focus on novel algorithms and representations for object detection and simplification of computer vision with neural networks. In particular the Lab plans to study the application of Graph Neural Networks for Object recognition and detection in images, videos and other signals, Combining information coming from various signals (multi-modal fusion) ant others. The lab research will be inspired by industry oriented problems.

deliverables
teaching courses on graph neural networks and computer vision
4 papers in the leading machine learning conferences, such as NeurIPS, CVPR, ICML, ICLR, etc.
open-source tools for working with graph neural neural networks
participation in 2 competitions on machine learning for graphs.
releasing a dataset for graph modeling in collaboration with other Skoltech groups
organizing one workshop on graph machine learning at one of the leading ML conferences