Papers
3
Total Citations
28
H-Index
3
About
Michel Sarkis is a computer vision researcher whose work centers on solving fundamental geometric problems in camera motion estimation and calibration. His most influential contribution is a robust solution to the five-point relative pose problem, a critical challenge for applications like robot navigation and telepresence. In his highly cited 2007 paper, he introduced a novel approach using Gauss-Newton optimization on a manifold, achieving fast and stable estimation of the essential matrix’s five degrees of freedom. This work, with 15 citations, provides a practical alternative to traditional methods. Sarkis also advanced camera calibration by modeling intrinsic parameter variation in automatic zoom systems using moving least-squares, a technique that improves accuracy in dynamic imaging conditions. His 2007 paper on this topic (9 citations) addresses the need for precise parameter estimates in real-world machine vision. Extending his manifold-based optimization to humanoid pose estimation, he demonstrated its versatility in stereo camera setups for telepresence robots. Sarkis’s contributions are foundational for researchers developing robust, real-time vision systems for autonomous platforms.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3