Skip to main content
Usami Laboratory

Department of Computer Science, College of Engineering, Chubu University

Usami Laboratory

Understanding the world from images

We study computer vision, medical image analysis, and the evaluation and reliability of large language models, connecting fundamental research with medicine, manufacturing, and social infrastructure.

Research Areas

Our four research areas cover physics-based image understanding, medical image analysis, spatiotemporal modeling, and generative AI and LLM evaluation.

Theme 01

Medical Image Analysis, 3D Shape Reconstruction, and Diagnostic Support

Research on lesion detection, segmentation, 3D shape reconstruction, quantitative measurement, and diagnostic support using endoscopy, CT, cellular, and pathological images.

View details
Theme 02

Physics-based Vision

Research on estimating shape, material, reflectance, and illumination from images by modeling reflection, transmission, scattering, and other optical phenomena.

View details
Theme 03

Real-world Sensing and Spatiotemporal Modeling

Research on extracting temporal and spatial structure from videos and sensor data for action recognition, forecasting, and anomaly detection.

View details
Theme 04

Generative AI, Learning Support, and Validation

Research on generative models, learning support, data augmentation, automatic question generation, and validation of generated outputs and LLM evaluators.

View details
View research overview

Selected Research

Selected research

The questions, methods, and evaluation results behind selected studies.

LLM Evaluation & Reliability

Testing LLM Judge Reliability

We test three properties of LLM judges: output validity, non-tie preferences between identical answers (dark current), and selection of a specified target.

LLM Judges Have Dark Current · arXiv preprint · 2026

Joint research with Mitsubishi Heavy Industries, Research & Innovation Center

Read research overview

Measurement stages

Output validity
Can the declared parser read the verdict?
Dark current
Does the judge prefer one exact duplicate?
Target-selection utility
Does it select the specified target?
01 / MEDICAL IMAGING

Anatomy-Preserving Augmentation via Generative AI

A workflow for augmenting medical training images with anatomical controls and generated-sample selection.

Explore the research02 / PHYSICS-BASED VISION

Separating Shape and Reflectance from a Single Image

A self-supervised method estimating scene properties like reflectance and shape from a single photo using relighting constraints.

Explore the research03 / SPATIOTEMPORAL MODELING

Building IoT Prediction with Sparse Graphs and Physical Constraints

Spatiotemporal prediction that limits graph neighbors and incorporates bandwidth and connection constraints into training.

Explore the research
All research explainers

Recent research outputs

The latest papers and presentations, including accepted and forthcoming work.

From Candidate Admission to Task-Level Return: Auditing LLM-Generated Code under Execution Contracts

Preprint · 2026

Usami, Hiroyasu, Hara, Keisuke, Tsuboi, Ayato, Matsuda, Naohiko

arXiv · arXiv:2610.24609; 34-page main text and 7-page ancillary material

Publication record

Physics-Guided Monocular Depth Estimation in Colonoscopy via Privileged Illumination Information

International Conference · 2026

Hara, Keisuke, Tsuboi, Ayato, Fukui, Shinji, Kato, Shunsuke, Ono, Satoshi, Takahama, Takuya, Ebi, Masahide, Ogasawara, Naotaka, Kasugai, Kunio, Usami, Hiroyasu

2026 19th International Conference on Machine Vision (ICMV 2026) · Presented at ICMV 2026; Paper ID M045; conference held October 15-18, 2026, Budapest, Hungary

Publication record

Dynamic BGP Route Control Based on Real-Time Performance Information Using an LLM with a Verification Mechanism

Domestic Conference · 2026

Combines LLM-based decisions with a verification mechanism to study network route control using real-time performance information.

Tsunogai, Kento, Yamasaki, Shugo, Kawabe, Kaito, Usami, Hiroyasu

Proceedings of the 2026 Tokai-Section Joint Conference on Electrical, Electronics, Information, and Related Engineering · 2B-2-7, p. 154, Network Systems, August 30, 2026

Publication record
View all publications

Laboratory record

Research outputs and awards featured on this site since the laboratory was founded in April 2022.

Hiroyasu Usami — profile

News

Research presentations, student activities, awards, and academic service.

Research and laboratory activities

Presentation · Generative AI

Dark Current code and supporting records are now available on GitHub

We have released the code and supporting records for the revised LLM Judges Have Dark Current study, conducted with Mitsubishi Heavy Industries, Research & Innovation Center. The repository includes the manuscript, recorded evaluation results, analysis code, and verification and reproduction instructions.

Academic service

Academic Service

Joined the Editorial Board of Scientific Reports

Hiroyasu Usami joined the Editorial Board of Scientific Reports, a Nature Portfolio journal, in September 2026, with a focus on computer vision, machine learning, and medical image analysis.

View all activity

People

Faculty, graduate students, and undergraduates work together on image understanding and AI.

1

Faculty

25

Current Students

21

Alumni

View people

Selected Career Paths

Companies are listed in Japanese alphabetical order.

  • AISIN Digital Engineering Co., Ltd.
  • Aiphone Co., Ltd.
  • ARGO GRAPHICS Inc.
  • Enshu Limited
  • Gothic Co., Ltd.
  • Common Products Co., Ltd.
  • C-Cube Co., Ltd.
  • J:COM Co., Ltd.
  • JBCC Corporation
  • Sumitomo Riko Company Limited
  • TechPresso Inc.
  • DIC Solution Engineering Co., Ltd.
  • Nodajuku Co., Ltd.

Graduate study: Chubu University Graduate School

For Prospective Students

Research topics, joining procedures, and information on laboratory visits and consultations.

View joining information

For Research Collaboration

Contact us about collaborative research in image analysis, machine learning, and LLM evaluation.

Contact us