Tomoya Fujii

Papers

Total Citations

142

H-Index

About

Tomoya Fujii is a researcher at the intersection of computer vision and bioinspired robotics, whose work spans from efficient deep learning to the mechanics of flapping-wing flight. His most cited contribution, "A Lightweight YOLOv2" (2018, 135 citations), addresses the critical challenge of real-time object detection in embedded systems—such as robotics and autonomous driving—by reformulating frame-based detection into a combined regression and classification problem optimized for high-speed, low-resource environments. This work has become a foundational reference for deploying neural networks on edge devices. More recently, Fujii has turned to biomimetics, exploring how hummingbirds and bats achieve complex torsional wing deformations during hovering flight. In "Hummingbird-bat hybrid wing by 3-D printing" (2023), he pioneered the use of additive manufacturing to replicate these natural wing structures, enabling unprecedented design control over passive deformation. His follow-up study on "Soft Limitation of Passive Feathering at Wing Root" (2023) introduced the "connecting membrane"—a novel elastic structure that improves lift generation by passively constraining wing root motion. Though early in citation impact, these works represent a novel synthesis of materials science, aerodynamics, and robotics, positioning Fujii as a creative force in both efficient vision systems and next-generation aerial robot design.