Takuma Yagi

The University of Tokyo

Papers

1

Total Citations

12

H-Index

1

About

Takuma Yagi is a computer vision researcher whose work centers on egocentric video understanding, hand-object interaction, and action anticipation. His research sits at the intersection of human activity recognition and robot learning, with a particular focus on how machines can interpret and predict human behavior from a first-person perspective. One of his most notable contributions is the development of fine-grained affordance annotation frameworks for egocentric hand-object interaction videos, a work that has garnered 12 citations since its publication in 2023. This research addresses a fundamental challenge in the field: rigorously defining and annotating object affordances — the action possibilities determined by human motor capacity and an object's physical properties — to advance tasks such as action anticipation and robot imitation learning. By establishing clearer annotation protocols and conceptual frameworks around affordance in egocentric contexts, Yagi's work provides essential groundwork for training more capable AI systems that can understand and replicate human interactions with objects. His contributions are increasingly relevant as the robotics and embodied AI communities seek richer, semantically grounded video datasets to bridge the gap between human demonstration and machine learning.

Research Focus

Key Achievements

1
H-Index
1
Papers
12
Total Citations
12
Avg Citations/Paper
🏆 Most Cited Paper
Fine-grained Affordance Annotation for Egocentric Hand-Object Interaction Videos
12 citations · 2023
📈 Most Prolific Year: 2023 (1 Papers)
🤝 Key Collaborators: 5
🏛 Institutions: The University of Tokyo

Top Papers

  1. 1

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago