Kazuto Nakashima
Associate Professor
Fukuoka Prefecture, Japan
Summary
🤩
Rockstar🎓
Top SchoolKazuto Nakashima is an associate professor and machine learning researcher based in Fukuoka with 11 years of experience bridging academia and applied AI. He holds a PhD in Information Science and Electrical Engineering from Kyushu University and progressed through research roles there to his current faculty position. His work includes hands-on deep learning engineering—contributing fixes and ResNet/classifier implementations to a well-known DeepLab v2 PyTorch repository for semantic segmentation. He has research experience at institutions including NASA JPL and was a JSPS research fellow, showing a mix of international collaboration and national research funding recognition. Colleagues describe him as someone who moves fluently between theory and reproducible code, often improving training and data-loading pipelines behind the scenes. He combines rigorous academic publication skills with practical open-source contributions that drive reproducible computer-vision research.
11 years of coding experience
5 years of employment as a software developer
Associate’s Degree, Electronics and Control Engineering, Associate’s Degree, Electronics and Control Engineering at Kumamoto National College of TechnologyDoctor of Philosophy - PhD, Information Science and Electrical Engineering, Doctor of Philosophy - PhD, Information Science and Electrical Engineering at Kyushu University
Japanese, English
Github Skills (5)
semantic-segmentation10
computer-vision10
pytorch10
resnet10
python10
Programming languages (2)
Github contributions (5)
kazuto1011/deeplab-pytorch
Nov 2017 - Aug 2020
PyTorch implementation of DeepLab v2 on COCO-Stuff / PASCAL VOC
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 101 commits, 29 PRs in 2 years 9 months
Contributions summary:Kazuto primarily contributed to the DeepLab-Pytorch repository, which implements DeepLab v2 for semantic segmentation. Their work involved implementing and fixing aspects of the DeepLab model, specifically focusing on the structure of the ResNet backbone and the classifier module. They addressed bugs in training scripts and data loading, and made changes to the demonstration scripts.
pytorchpascalpascal-vocsemantic-segmentationdeeplab
kazuto1011/r2dm
Sep 2023 - Jan 2025
LiDAR Data Synthesis with Denoising Diffusion Probabilistic Models (ICRA 2024)
Contributions:1 release, 4 PRs, 19 pushes in 1 year 3 months
diffusion-modelslidar
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.