Taiyu Long
Papers
1
Total Citations
3
H-Index
1
About
Taiyu Long is a researcher advancing the frontier of video understanding, with a primary focus on efficient spatio-temporal action detection. His key contributions center on developing energy-efficient frameworks for human-centric video analysis, a domain critical for applications in robotics, security, and healthcare. Long’s notable work, "E²TAD: An Energy-Efficient Tracking-based Action Detector" (2022), introduces a novel paradigm that reimagines the two-stage detection pipeline—inspired by Faster R-CNN—by leveraging tracking mechanisms to dramatically reduce computational overhead. This approach enables real-time, accurate localization of human actions across both space and time without sacrificing performance. While still early in his career, with 3 citations on this flagship paper, Long’s work addresses a pressing need for sustainable AI in video analytics, balancing high accuracy with low energy consumption. His research is particularly impactful for deploying action detection on resource-constrained devices, such as edge computing systems in autonomous robots or surveillance cameras. By tackling the energy bottleneck in video understanding, Taiyu Long is paving the way for more practical and scalable human-centric intelligent systems, making him a promising voice in the next generation of computer vision researchers.
Research Focus
Key Achievements
Top Papers
- 1E^2TAD: An Energy-Efficient Tracking-based Action Detector3 citations · 2022