Muhammad Arslan Manzoor

Nanjing University of Science and Technology

Papers

Total Citations

H-Index

About

Dr. Muhammad Arslan Manzoor is a computer vision researcher whose work focuses on making deep neural networks practical for real-time applications. His primary research areas include semantic segmentation, efficient neural network architectures, and real-time image processing. Dr. Manzoor’s most notable contribution is the development of SPSSNet, a real-time network for image semantic segmentation that addresses the critical challenge of balancing accuracy with computational efficiency. While many deep neural networks suffer from sluggish performance due to excessive feature channels, parameters, and floating-point operations, SPSSNet was designed to deliver high-quality segmentation without the heavy computational burden, making it suitable for time-sensitive applications. This work has garnered attention in the field, with his most-cited paper accumulating 9 citations and establishing him as a researcher dedicated to bridging the gap between state-of-the-art deep learning and practical deployment. Dr. Manzoor’s research is particularly valuable for students and engineers working on autonomous systems, robotics, and mobile vision, where real-time performance is essential. His contributions continue to inspire efficient model design in the growing domain of edge AI and embedded vision systems.