Kourosh Khoshelham

Papers

Total Citations

H-Index

About

Kourosh Khoshelham is a leading figure in geospatial and computer vision research, whose work bridges the gap between 3D spatial data and intelligent perception. His primary research areas include point cloud processing, indoor navigation, and self-supervised depth estimation. A major contribution is his development of obstacle-aware indoor pathfinding using point clouds, a method that extracts navigable spaces from 3D scans to support applications like assistive navigation for the disabled and emergency response. This work, published in 2019, has garnered 20 citations for its practical impact on urban spatial information. More recently, Khoshelham has advanced self-supervised monocular depth estimation with a novel joint attention mechanism and intelligent mask loss (2024, 3 citations), improving depth prediction without labeled data—a key step toward robust autonomous systems. His research is notable for its direct relevance to real-world challenges, from robotics to augmented reality, and his ability to integrate geometric reasoning with deep learning. With a citation record reflecting steady influence, Khoshelham continues to shape how machines understand and navigate complex indoor environments.