Boyuan Feng
Papers
1
Total Citations
2
H-Index
1
About
Boyuan Feng is a rising researcher in the field of efficient deep learning and video processing systems. His work focuses on bridging the gap between the high computational demands of convolutional neural networks (CNNs) and the resource constraints of cloud-backed mobile platforms, including cell phones and robotics. Feng’s major contribution lies in addressing the critical challenge of dynamic class skew in real-time video analytics—a scenario where the distribution of object classes shifts unpredictably during processing. His notable work, "Palleon: A Runtime System for Efficient Video Processing toward Dynamic Class Skew" (2021), introduces a novel runtime system that intelligently allocates computational resources to maintain high accuracy while minimizing energy consumption. Although early in its citation impact, this work has been recognized for tackling a practical bottleneck in deploying CNNs on mobile devices. Feng’s research is pivotal for enabling sustainable, real-time AI applications, and his innovative approach to runtime optimization positions him as a promising contributor to the future of edge AI and energy-efficient video analytics.
Research Focus
Key Achievements
Top Papers
- 1