Sergey Zakharov

Papers

Total Citations

H-Index

About

Sergey Zakharov is a researcher working at the intersection of computer vision, robotics, and neural scene representation. His work focuses on developing intelligent systems capable of understanding and interacting with complex visual environments, with a particular emphasis on object-centric learning and robotic manipulation. His most notable contribution, "Multi-Object Manipulation via Object-Centric Neural Scattering Functions" (2023), addresses one of the fundamental challenges in robotic manipulation: how to effectively represent and reason about scenes containing multiple interacting objects. By leveraging neural scattering functions within an object-centric framework, Zakharov's approach advances beyond conventional scene decomposition methods that struggle with precise modeling under real-world complexity. This work sits at a critical frontier where visual perception meets physical interaction, enabling robots to better plan and execute manipulation tasks in cluttered environments. While his citation count is still growing — reflecting the recency of his contributions — the problems he tackles are central to the future of autonomous robotics and embodied AI. Researchers working on visual dynamics modeling, sim-to-real transfer, or object-level scene understanding will find his work particularly relevant and forward-looking.