Papers

41

Total Citations

490

H-Index

13

About

Wataru Takano is a pioneering robotics researcher whose work sits at the intersection of human motion understanding, natural language processing, and humanoid robot intelligence. His research has fundamentally advanced how robots perceive, symbolize, and generate human movement, enabling richer forms of human-robot communication. Takano's most influential contributions center on bridging the gap between physical motion and language. His hierarchical mimesis model for primitive nonverbal communication (2006, 52 citations) established an early framework for social robots capable of interpreting and mimicking human behavior. Building on this, he developed statistical models enabling bidirectional translation between whole-body motion primitives and natural language sentences (2015, 46 citations), allowing robots to both describe movements linguistically and generate motion from verbal instructions. A recurring theme across his body of work is unsupervised segmentation of continuous human motion into discrete, meaningful primitives — a capability critical for real-time robot learning. His later deep learning approaches, including the HVGH framework (2019), extend these methods to high-dimensional data using neural compression. More recently, his IMU-based annotation generation work (2020) demonstrates a commitment to practical, real-world deployment. Collectively, Takano's research has shaped how humanoid robots acquire, interpret, and communicate knowledge about human movement.

Research Focus

Key Achievements

13
H-Index
41
Papers
490
Total Citations
12
Avg Citations/Paper
🏆 Most Cited Paper
Primitive communication based on motion recognition and generation with hierarchical mimesis model
52 citations · 2006
📈 Most Prolific Year: 2015 (12 Papers)
🤝 Key Collaborators: 26
🏛 Institutions: The University of Tokyo, The University of Osaka, Bunkyo University, Toyo University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago