Ilya A. Petrov

Papers

Total Citations

152

H-Index

About

Ilya A. Petrov is a computer vision and machine learning researcher whose work centers on understanding and modeling human behavior through the lens of human-object interactions. His most recognized contribution is the BEHAVE dataset and methodology, a landmark resource designed to advance the tracking of interactions between people and objects in natural, unconstrained environments. This work addresses one of the field's most demanding challenges: enabling systems to generalize across a vast range of objects and interaction scenarios, with direct applications in gaming, virtual and mixed reality, human behavior analysis, and human-robot collaboration. The BEHAVE framework has garnered significant attention from the research community, accumulating over 150 citations since its publication in 2022, reflecting the genuine need it fulfills for researchers working on embodied AI and 3D scene understanding. By providing both a richly annotated dataset and a principled tracking method, Petrov and his collaborators have handed the community a practical foundation upon which more robust, real-world interaction systems can be built. His research sits at an important intersection of computer vision, human motion modeling, and applied AI, making his contributions highly relevant to both academic researchers and industry practitioners pushing the boundaries of immersive and collaborative technologies.