Papers
6
Total Citations
136
H-Index
4
About
Qingyi Gu is a leading researcher at the intersection of computer vision, robotics, and deep learning, with a particular focus on high-speed vision systems and real-time 3D perception. His work addresses the critical challenge of deploying sophisticated visual intelligence on resource-constrained platforms like mobile robots and smart manufacturing systems. Gu’s most influential contribution is his pioneering work on network compression, notably his 2018 paper on a novel channel pruning method for deep neural networks (67 citations), which provides a practical pathway for running complex models on embedded devices without sacrificing performance. He has also made significant advances in high-speed vision, developing a robot-mounted structured light system capable of capturing 3D shape measurements at 500 fps, enabling precise inspection of moving objects. In robotics, Gu has explored dexterous manipulation, such as realizing flower stick rotation with a robotic arm, demonstrating feedback control strategies for complex dynamic tasks. His research extends to wide-area action analysis and 6-D pose estimation using efficient sampling methods, showcasing his commitment to bridging the gap between theoretical algorithms and real-world, high-speed applications in security and industrial automation.
Research Focus
Key Achievements
Top Papers
- 1A novel channel pruning method for deep neural network compression67 citations · 2018
- 2
- 3
- 4Realization of flower stick rotation using robotic arm14 citations · 2015
- 5
- 6