Fahad Shahbaz Khan

Linköping University, Cranfield University

Papers

Total Citations

298

H-Index

About

Fahad Shahbaz Khan is a leading researcher in computer vision and robotics, with a primary focus on enabling deep learning systems to operate effectively on sparse and irregularly structured data—a critical challenge for real-world applications like autonomous driving, robotics, and surveillance. His most impactful contributions center on the development of confidence propagation mechanisms within convolutional neural networks (CNNs). In his highly cited 2019 work, "Confidence Propagation through CNNs for Guided Sparse Depth Regression" (221 citations), Khan introduced a novel framework that allows CNNs to process sparse depth maps by propagating confidence scores through the network, dramatically improving depth estimation from limited sensor data. This work builds on his foundational 2018 paper (75 citations), which first established the principles of confidence-guided regression for irregularly spaced inputs. Beyond theoretical advances, Khan demonstrates a strong commitment to practical deployment, as evidenced by his recent 2024 work on a ROS 2-based software framework for industrial object handling. His research bridges the gap between algorithmic innovation and real-world robotic systems, making him a pivotal figure in the advancement of perception-driven automation.