Hao-Yu Miao
Papers
1
Total Citations
3
H-Index
1
About
Hao-Yu Miao is a researcher focused on energy-efficient computer vision, particularly in video action detection and spatio-temporal action localization. His work addresses the critical challenge of deploying intelligent video analysis—essential for robotics, security, and healthcare—on resource-constrained devices. Miao’s major contribution, the E²TAD (Energy-Efficient Tracking-based Action Detector), introduces a novel two-stage paradigm inspired by Faster R-CNN, optimizing the balance between detection accuracy and computational efficiency. This approach has garnered attention in the field, with his most-cited paper accumulating 3 citations since 2022. By prioritizing energy efficiency without sacrificing performance, Miao’s research paves the way for practical, real-time human-centric analysis in edge computing environments. His work stands out for its innovative integration of tracking mechanisms into action detection, reducing power consumption while maintaining robust spatio-temporal localization. As the demand for sustainable AI grows, Miao’s contributions are poised to influence future developments in efficient video understanding, making him a notable emerging voice in computer vision.
Research Focus
Key Achievements
Top Papers
- 1E^2TAD: An Energy-Efficient Tracking-based Action Detector3 citations · 2022