Mohammad Nomaan Qureshi
Papers
1
Total Citations
6
H-Index
1
About
Mohammad Nomaan Qureshi is a researcher advancing the frontier of robot learning from visual demonstrations. His primary research areas lie at the intersection of computer vision, robotics, and physics-based simulation, with a focus on enabling machines to acquire complex manipulation skills through observation. In his notable 2022 work, "Learning Object Manipulation Skills from Video via Approximate Differentiable Physics," Qureshi introduced a novel optimization framework that allows robots to learn simple object manipulation tasks from a single video demonstration. By reconstructing a coarse, temporally evolving 3D scene that mimics the demonstrated action, his approach bypasses the need for extensive training data or manual programming. This work, which has garnered 6 citations, represents a significant step toward more autonomous and data-efficient robot learning. Qureshi’s contributions are particularly impactful for researchers in imitation learning and physical reasoning, offering a scalable path to transfer human skills to robotic systems. His innovative use of differentiable physics to bridge the gap between 2D video and 3D action execution marks him as a promising voice in embodied AI.
Research Focus
Key Achievements
Top Papers
- 1