Te-Lin Wu

Papers

Total Citations

108

H-Index

About

Te-Lin Wu is a researcher whose work sits at the intersection of computer vision, robotics, and cognitive science, with a particular focus on enabling machines to understand and reason about object affordances—the potential actions that objects offer. His most influential contribution is the development of the Demo2Vec model, introduced in a 2018 paper that has garnered 108 citations. This work addresses a fundamental challenge in robotics: how can a robot learn what it can do with an unfamiliar object simply by watching a human demonstration? By learning to extract feature embeddings from demonstration videos, Demo2Vec allows machines to generalize affordance knowledge to unseen objects, bridging the gap between human-like observational learning and artificial intelligence. This research has significant implications for autonomous systems, making robots more adaptable and intuitive in real-world environments. Wu’s work is notable for its elegant fusion of deep learning and cognitive principles, offering a practical pathway toward more intelligent, perceptive machines that learn from natural human behavior.