Maohui Li
Papers
1
Total Citations
7
H-Index
1
About
Maohui Li is a leading researcher in efficient computer vision, specializing in instance semantic segmentation and knowledge distillation for transformer-based architectures. His most cited work, "Knowledge Distillation for Efficient Instance Semantic Segmentation with Transformers" (2024, 7 citations), addresses a critical challenge in modern AI: the computational burden of state-of-the-art models like Mask2Former. Li’s key contribution lies in developing a knowledge distillation framework that compresses these complex models—which provide detailed per-pixel scene understanding for computer vision and robotics—without sacrificing accuracy. By transferring knowledge from a cumbersome teacher network to a lightweight student, his approach enables real-time, high-quality segmentation on resource-constrained devices. This work bridges the gap between cutting-edge performance and practical deployment, making Li a pivotal figure in advancing efficient deep learning. His research has immediate implications for autonomous systems, augmented reality, and edge computing, where low-latency scene understanding is essential. With a growing citation impact and a focus on sustainable AI, Maohui Li continues to shape the future of efficient visual perception.
Research Focus
Key Achievements
Top Papers
- 1