Naoya Fushishita
Papers
1
Total Citations
2
H-Index
1
About
Naoya Fushishita is a researcher focused on computer vision and video prediction, with a particular emphasis on generating long-term, multi-future scenarios from input videos. His work addresses a critical limitation in autonomous driving and robotics: predicting not just one, but multiple plausible futures to account for diverse behavioral outcomes. His most-cited paper, "Long-Term Video Generation of Multiple Futures Using Human Poses" (2019), introduces a novel framework that leverages human pose information to extend prediction horizons, enabling more robust and realistic forecasting. While still emerging, with 2 citations to date, this contribution lays foundational groundwork for safer, more adaptive AI systems in dynamic environments. Fushishita’s research bridges the gap between short-term single-path predictions and the complex, branching realities of real-world decision-making, marking him as an innovator in predictive vision technology.
Research Focus
Key Achievements
Top Papers
- 1Long-Term Video Generation of Multiple Futures Using Human Poses.2 citations · 2019