Mohsen Ghafoorian

Sharif University of Technology, Qualcomm (United Kingdom)

Papers

Total Citations

H-Index

About

Mohsen Ghafoorian’s research lies at the intersection of reinforcement learning and computer vision, with a focus on enabling intelligent systems to perceive, learn, and act autonomously. His early work introduced a novel approach to automatic abstraction in reinforcement learning using an ant system algorithm, a contribution that has garnered foundational recognition with 9 citations. More recently, Ghafoorian has advanced the field of 3D scene understanding through FastCAD, a real-time system for CAD model retrieval and alignment from scans and videos, which has already earned 6 citations since its 2024 publication. This work demonstrates his ability to bridge theoretical learning methods with practical, real-world applications—particularly in robotics, autonomous navigation, and augmented reality. By tackling the challenge of efficient object recognition and pose estimation from dynamic visual data, Ghafoorian’s research offers scalable solutions for interactive AI systems. His trajectory from foundational reinforcement learning algorithms to cutting-edge computer vision pipelines highlights a commitment to developing autonomous agents that can robustly interpret and act within complex environments.