Te-Cheng Liu

Papers

Total Citations

H-Index

About

Te-Cheng Liu is a researcher whose work lies at the intersection of computer vision and human-robot interaction, with a particular focus on enabling machines to understand human body language. His key research areas include hand posture recognition, gesture analysis, and the application of probabilistic graphical models to visual perception. Liu’s most cited work, "Hand posture recognition using Hidden Conditional Random Fields" (2009), addresses a critical challenge in human-robot interaction: robustly interpreting hand gestures in real-world environments. By leveraging local features like SIFT within a Hidden Conditional Random Field framework, his approach achieves invariance to cluttered backgrounds—a significant step toward natural, non-verbal communication between humans and robots. This foundational paper has garnered 7 citations, reflecting its influence on subsequent gesture recognition systems. Liu’s contributions are particularly notable for bridging the gap between theoretical machine learning models and practical robotic applications, demonstrating how structured prediction can enhance perceptual robustness. For students and researchers exploring human-robot interaction, Liu’s work offers a clear example of how combining feature engineering with probabilistic reasoning can solve real-world perception problems, making body-language understanding more reliable and context-aware.