Papers
1
Total Citations
7
H-Index
1
About
Amr Bakry is a computer vision researcher whose work bridges object recognition and pose estimation—two foundational challenges for intelligent vision and robotic systems. His most-cited paper, "Joint object recognition and pose estimation using a nonlinear view-invariant latent generative model" (2016, 7 citations), introduces a unified framework that models how multiple viewpoints of an object lie on an intrinsic low-dimensional manifold in the input space. By learning a nonlinear, view-invariant latent representation, Bakry’s approach enables simultaneous identification of an object and its viewpoint, overcoming the traditional separation of these tasks. This joint modeling is particularly valuable for applications in autonomous navigation, manipulation, and scene understanding, where recognizing both what an object is and how it is oriented is critical. Bakry’s work contributes to the broader goal of building robust, viewpoint-adaptive vision systems, and his research continues to influence developments in generative models for visual perception.
Research Focus
Key Achievements
Top Papers
- 1