Kojiro Takeyama

About

Kojiro Takeyama is a rising researcher at the forefront of human-centered AI, with a primary focus on integrating trajectory data and large language models (LLMs) for scene-aware human action prediction. His most-cited work, "TR-LLM: Integrating Trajectory Data for Scene-Aware LLM-Based Human Action Prediction" (2025), addresses a critical challenge in autonomous systems: accurately anticipating human behavior under real-world constraints like occlusions and incomplete scene observations. By fusing trajectory information with LLM reasoning, Takeyama’s approach enables robots and AI agents to infer intent and predict actions even when visual data is partial or ambiguous. This work has already garnered early citations, signaling its impact on the intersection of computer vision, robotics, and natural language processing. Takeyama’s contributions are particularly notable for bridging the gap between low-level motion tracking and high-level semantic understanding, offering a scalable framework for assistive technologies. As a young researcher, his work is poised to influence the next generation of context-aware AI, with potential applications in autonomous navigation, human-robot collaboration, and smart environments.

Research Focus

Key Achievements

Top Papers

1. 1
   TR-LLM: Integrating Trajectory Data for Scene-Aware LLM-Based Human Action Prediction

   3 citations · 2025

Key Collaborators

Contact & Links