Tomoya Ohta
Papers
1
Total Citations
7
H-Index
1
About
Tomoya Ohta’s research lies at the intersection of visual place recognition, scene understanding, and robotic perception. His most cited work, “Scene graph descriptors for visual place classification from noisy scene data” (2023, 7 citations), pioneers the use of scene graphs—rich models capturing appearance, spatial, and semantic relationships—to classify visual places despite noisy or incomplete data. This contribution addresses a critical challenge in robotics: enabling machines to robustly recognize environments by leveraging complex contextual cues rather than isolated features. Ohta’s approach advances how autonomous systems interpret dynamic scenes, with implications for navigation, mapping, and human-robot interaction. His work is notable for bridging graph-based scene representation with practical classification under real-world noise, a step toward more resilient visual place recognition. By focusing on the structural and relational aspects of scenes, Ohta offers a fresh perspective in a field often dominated by deep learning alone. His research continues to inspire new methods for integrating context into robotic perception, making him a promising voice in the growing domain of scene-aware robotics.
Research Focus
Key Achievements
Top Papers
- 1