Akira Jinguji

Tokyo Institute of Technology, National Institute of Information and Communications Technology

Papers

Total Citations

H-Index

About

Akira Jinguji is a researcher specializing in embedded deep learning, FPGA-based hardware acceleration, and efficient neural network deployment for real-time computer vision applications. His work sits at the intersection of machine learning and embedded systems, with a particular focus on making computationally demanding AI tasks feasible on resource-constrained hardware platforms. Jinguji's most notable contribution is his 2020 paper on fast monocular depth estimation for FPGAs, which has garnered 8 citations and addresses the critical challenge of inferring pixel-wise 3D scene depth from a single general-purpose camera. This work has direct implications for robotics, autonomous vehicles, and drone navigation, where low-cost, low-power depth sensing is essential. His subsequent 2021 research on weight sparseness for feature-map-split CNNs further advances the field by enabling efficient image recognition on low-end FPGAs through network compression techniques, accumulating 3 citations. Together, these contributions reflect a coherent research agenda aimed at democratizing AI inference on affordable embedded platforms. Jinguji's work is particularly relevant for students and engineers exploring edge AI, offering practical algorithmic and architectural strategies for deploying neural networks where computational resources are severely limited.