S. M. Reza Soroushmehr

Papers

Total Citations

H-Index

About

S. M. Reza Soroushmehr is a researcher whose work lies at the intersection of computer vision and deep learning, with a particular focus on single image depth estimation—a fundamental challenge for applications ranging from 3D modeling to autonomous robotics. His major contributions center on developing novel frameworks that effectively combine local and global image features to solve the ill-posed problem of inferring depth from a single monocular image. Notably, his 2018 paper, "Aggregation of Rich Depth-Aware Features in a Modified Stacked Generalization Model for Single Image Depth Estimation," which has garnered 23 citations, introduces an innovative stacked generalization approach that aggregates rich depth-aware features, demonstrating a sophisticated method for improving depth prediction accuracy. This work, along with his earlier 2016 study on joint local-global features, showcases his commitment to advancing the field by addressing the inherent ambiguity in monocular depth estimation. Soroushmehr’s research is particularly impactful for enabling practical applications such as 2D-to-3D conversion and robot vision, making his contributions valuable for both academic researchers and industry practitioners seeking robust depth perception solutions.