Sung Jae Lee
Papers
2
Total Citations
9
H-Index
2
About
Sung Jae Lee is advancing precision agriculture through cutting-edge deep learning and computer vision. His research focuses on two critical challenges: high-fidelity image reconstruction and robust semantic segmentation for crop and weed management. Lee’s major contributions include the development of **CNCAN** (Contrast and Normal Channel Attention Network), a novel architecture for super-resolution image reconstruction that enables cost-effective, high-resolution imaging from low-cost cameras. Building on this, he introduced **KDOSS-net**, a knowledge distillation-based outpainting and semantic segmentation network that achieves pixel-level classification of crops, weeds, and background with remarkable efficiency. These works directly address the practical constraints of smart farming, where affordable hardware often limits performance. With his most-cited papers accumulating early citations (5 and 4 respectively), Lee’s impact is already evident in the agricultural AI community. His notable achievement lies in bridging the gap between theoretical model design and real-world deployment, offering scalable solutions for robot-based agriculture. By integrating attention mechanisms and knowledge distillation, Lee is paving the way for more accessible, accurate, and automated weed management systems—a cornerstone for sustainable crop yield improvement.
Research Focus
Key Achievements
Top Papers
- 1
- 2