Mengquan Li
Papers
1
Total Citations
4
H-Index
1
About
Mengquan Li is a researcher at the forefront of efficient 3D computer vision hardware, specializing in point cloud neural network acceleration. His work addresses the critical challenge of deploying complex 3D perception models—essential for autonomous driving, robotics, and virtual reality—on resource-constrained edge devices. Li’s major contribution is the development of novel architectural and algorithmic co-designs that dramatically improve the energy and time efficiency of point cloud accelerators. His most-cited paper, “SimDiff: Point Cloud Acceleration by Utilizing Spatial Similarity and Differential Execution” (2024, 4 citations), introduces a pioneering technique that exploits spatial redundancy in point cloud data to reduce unnecessary computations. By leveraging spatial similarity and differential execution, SimDiff achieves significant performance gains over prior accelerators, marking a notable step toward real-time, low-power 3D perception. Li’s work is highly relevant for students and researchers interested in the intersection of computer architecture, embedded systems, and 3D vision, offering practical insights into building hardware that can keep pace with the growing demands of autonomous systems and immersive technologies.
Research Focus
Key Achievements
Top Papers
- 1