Tomikazu Tanuki

Papers

Total Citations

H-Index

About

Tomikazu Tanuki’s research lies at the intersection of robotics, human-computer interaction, and cognitive modeling, with a core focus on how machines can learn from human demonstration. His pioneering work on **attention point analysis** offers a powerful framework for constructing symbolic task models from observed human behavior. In his most cited paper, “Recognition of human task by attention point analysis” (2002, 25 citations), Tanuki introduced a two-step method that first broadly observes a task to build a rough model, then identifies critical “attention points” requiring deeper analysis—dramatically improving efficiency in task recognition. He extended this approach in “Acquiring hand-action models by attention point analysis” (2002, 22 citations), where he demonstrated how robots can learn hand-action representations by integrating multiple observational cues. These contributions are foundational for **learning from demonstration** and **imitation learning**, enabling robots to acquire complex manipulation skills without explicit programming. Tanuki’s work has influenced subsequent research in autonomous robotics and human-robot collaboration, providing a principled method for bridging the gap between raw sensory data and high-level symbolic understanding. His attention point paradigm remains a valuable tool for researchers seeking to build more adaptive, perceptive robotic systems.