Shengcao Cao
Papers
1
Total Citations
3
H-Index
1
About
Shengcao Cao is a researcher advancing the frontiers of efficient deep learning, with a primary focus on knowledge distillation and lightweight model design for resource-constrained perception systems. His most-cited work, "Learning Lightweight Object Detectors via Multi-Teacher Progressive Distillation" (2023, 3 citations), tackles a critical challenge in edge computing and vision-for-robotics: achieving high accuracy in object detection while maintaining minimal computational and memory footprints. Building on the proven success of knowledge distillation for classification, Cao innovatively extends this paradigm to object detection by introducing a multi-teacher progressive distillation framework. This approach systematically transfers knowledge from multiple, more capable teacher models to a compact student detector, enabling significant performance gains without sacrificing efficiency. His work directly addresses the pressing need for deployable AI in real-world applications where hardware constraints are paramount. By pioneering methods that bridge the gap between accuracy and efficiency, Cao is making impactful contributions to the practical deployment of vision models, empowering everything from autonomous robots to edge devices with state-of-the-art perceptual capabilities.
Research Focus
Key Achievements
Top Papers
- 1