Mohammad Loni

Papers

Total Citations

H-Index

About

Mohammad Loni is a leading researcher in embedded artificial intelligence, with a primary focus on designing efficient deep learning architectures for autonomous systems. His work centers on the critical challenge of deploying complex neural networks on resource-constrained embedded platforms, particularly for real-time computer vision tasks. Loni’s most cited paper, “Designing Compact Convolutional Neural Network for Embedded Stereo Vision Systems” (2018, 15 citations), presents a groundbreaking method for creating lightweight CNNs that can process stereo vision data—essential for depth perception in applications ranging from autonomous surgery to self-driving cars. This contribution directly addresses the trade-off between computational efficiency and accuracy, enabling practical deployment of stereo vision in embedded environments. His research has significant implications for the safety and reliability of autonomous systems, as stereo cameras provide rich environmental data including depth, luminance, color, and shape. Beyond this flagship work, Loni’s broader portfolio explores neural architecture search and model compression techniques, consistently pushing the boundaries of what is achievable on low-power hardware. His work is widely recognized for bridging the gap between cutting-edge AI research and real-world embedded system constraints, making him a key figure in the advancement of autonomous robotics and edge intelligence.