Kayvan Najarian

Papers

Total Citations

H-Index

About

Kayvan Najarian is a leading researcher in computer vision and artificial intelligence, with a particular focus on depth estimation from single monocular images—a fundamental challenge for applications in robotics, 3D modeling, and 2D-to-3D conversion. His major contributions include pioneering methods that integrate local and global visual features to infer scene depth, addressing the ill-posed nature of the problem. Notably, his 2018 work on "Aggregation of Rich Depth-Aware Features in a Modified Stacked Generalization Model" (23 citations) introduced a novel ensemble approach that significantly improved depth prediction accuracy by combining multiple feature representations. His earlier 2016 study on "Single image depth estimation using joint local-global features" (3 citations) laid foundational groundwork for balancing fine-grained details with holistic scene context. Through these innovations, Najarian has advanced the capability of machines to perceive three-dimensional structure from flat images, enabling more robust autonomous navigation and spatial understanding. His research continues to influence the development of intelligent vision systems, making him a notable figure in the field of computational depth perception.