Wanjia Liu
Papers
1
Total Citations
7
H-Index
1
About
Wanjia Liu is a researcher whose work lies at the intersection of computer vision and efficient video understanding, with a particular focus on temporal representations and event-driven processing. Liu is best known for the paper "Fast Retinomorphic Event-Driven Representations for Video Gameplay and Action Recognition" (2019, 7 citations), which introduced a novel framework for capturing temporal dynamics in video data. This work proposed a retinomorphic approach that mimics biological vision systems to generate event-driven representations, offering a faster and more efficient alternative to traditional two-stream networks. By addressing the computational bottleneck of processing temporal information, Liu's contributions have implications for real-time action recognition and video gameplay analysis. While still early in their career, Liu's research demonstrates a commitment to pushing the boundaries of how machines perceive motion and time in visual data, laying groundwork for more responsive and biologically inspired AI systems.
Research Focus
Key Achievements
Top Papers
- 1