LAB LOGA page on LabLog, a review site (not the lab's official page)
Ikuro Sato Lab
Machine-translated · View original
A lab researching recognition technology that replaces human eyes, aimed at autonomous driving.
AboutAbout the lab
Replacing human visual recognition with machines.
The Ikuro Sato Lab was founded in April 2020 as a computer vision and machine learning lab.
The lab conducts research aimed at developing highly accurate recognition technology that can replace human visual recognition.
Centering on recognition that uses sensor information in autonomous driving and advanced driver assistance systems, it also covers anomaly detection and image processing in general.
Through a joint research course with DENSO IT Laboratory, Inc., the lab also began work on the area combining AI and control from fiscal year 2025.
Key terms
- Computer vision
- Technology that lets computers read what appears in images and video captured by cameras and other devices
- Backpropagation
- A method in neural network training that traces the error in the output backward from the output side to correct the weights
- Network quantization
- A technique that represents the numbers used inside a model at coarse levels to make computation lighter and faster
NewsLab news
Paper accepted to the English-language journal of the Institute of Electronics, Information and Communication Engineers
A researcher and a doctoral student received an Excellence Award and a Student Encouragement Award at MIRU2026
Paper accepted at the international conference BMVC 2026
Paper accepted at the international conference ECCV 2026
Paper accepted at the international conference ICML 2026
ResearchResearch themes
The lab develops highly accurate recognition technology with the aim of replacing human visual recognition with machines. While its main applications are recognition in autonomous driving and advanced driver assistance systems, it also pursues fundamental research that leads to a wider range of applications.
*Illustrative imageModels of memory
Modern Hopfield networks, known as a model of how memories are formed and recalled in the brain, have had remaining issues in how they behave on data from distributions they were not trained on. The lab aims to establish a method that resolves this issue from first principles.
*Illustrative imageExtracting general-purpose features
To replace driving with machines, recognition of pedestrians, vehicles, white lines, signs, and more must be handled at the same time. The lab develops an optimization method that suppresses co-adaptation between neurons and extracts better-generalized features.
*Illustrative imageMaking models compact
In safety-critical recognition such as autonomous driving, real-time processing is essential. To greatly speed up inference with minimal loss of performance, the lab develops techniques centered on network quantization.
*Illustrative imageTarget propagation
Backpropagation is said to differ in principle from learning in the brain in several respects. For target propagation, which has a certain degree of biological plausibility, the lab aims to solve the problem of learning instability with a simple method.
MembersFaculty
The Ikuro Sato Lab consists of faculty members, researchers, joint researchers from industry and other organizations, assistant staff, and doctoral and master's students.As of 2026
- Project ProfessorIkuro Ikuro SatoPublications ↗
- Project Associate ProfessorKenta HoshinoPublications ↗
VoicesStudent voices
Ratings and reviews reflect the experiences of individual current or former students, not the lab's official views. Whether a lab has reviews, or how many, says nothing about the lab itself.
Students who take part in an interview may receive a thank-you gift worth ¥1,000, regardless of what they say.
Student ratings (0 students)
--
--
--
--
--
--
Student voices answer questions like these
- What's great about this lab?
- What did you learn after joining?
- What's the vibe, and what is the professor like?
- Who is this lab for?
- Where do graduates go?
Your experience can help younger students choose their lab. Your name won't be shown.
Write a lab reviewAnonymous · Your draft is saved if you stopRelatedLabs similar to Ikuro Sato Lab
Asako Kanezaki LabSchool of Computing, Department of Computer ScienceComputer vision, data-driven intelligent robotics
Saito LabSchool of Computing, Department of Computer ScienceComputer graphics, image processing
Shimosaka LabSchool of Computing, Department of Computer ScienceUbiquitous computing and data analysis
Haruhiko Kaneko LabSchool of Computing, Department of Computer ScienceError control codes and data compression technologies (highl…The information on this page is compiled by the LabLog team from the lab's public information and interviews. For corrections or removal requests, please contact LabLog, not the lab.
Report an errorAbout the images
The top and research theme images are illustrative and were not taken at the lab. They are cropped from Wikimedia Commons images. Cropped CC BY-SA images are provided under the same CC BY-SA license as the original (version shown on each line).
- Top image: “Ouster OS1-64 lidar point cloud of intersection of Folsom and Dore St, San Francisco” Daniel L. Lu / CC BY 4.0 (cropped)
- Models of memory: “PNNs at primary somatosensory cortex in mouse brain 03” Арнст Никита, Светлана Кузнецова, Мельникова Анастасия / CC BY-SA 4.0 (cropped)
- Extracting general-purpose features: “Shibuya Crossing 20241020” This Photo was taken by Supanut Arunoprayote. Feel free to use any of my images, but please mention me as the author and may send me a message. (สามารถใช้ภาพได้อิสระ แต่กรุณาใส่เครดิตผู้ถ่ายและอาจส่งข้อความบอกกล่าวด้วย) Please do not upload an updated image here without consultation with the Author. The author would like to make corrections only at his own source. This ensures that the changes are preserved.Please if you think that any changes should be required, please inform the author.Otherwise you can upload a new image with a new name. Please use one of the templates derivative or extract. / CC BY 4.0 (cropped)
- Making models compact: “NVidia Jetson Orin AGX” Auledas / CC BY 4.0 (cropped)
- Target propagation: “Artificial Neural Network Example” Jhedengren / CC BY-SA 4.0 (cropped)
- Related lab (金崎研究室): “Atlas frontview 2013” DARPA / Public domain (cropped)
- Related lab (齋藤研究室): “Utah teapot surreal” Dhatfield / CC BY-SA 3.0 (cropped)
- Related lab (下坂研究室): “Warsaw city center in winter at night December 2023” Oleslawlama / CC BY-SA 4.0 (cropped)
- Related lab (金子研究室): “Hard disk head crash” Alchemist-hp / CC BY-SA 3.0 (cropped)
About LabLog
LabLog is a lab review app for students only. Student members at each university interview students in labs and publish what they say. Student voices can only be read by students who sign up with a university email address. Operated by bestiee Inc.
Learn more about LabLog ↗Search engines and AI may reference the basic lab info and rating summaries on this page and the description of LabLog. Student voices are for student members only. Automated collection by programs (scraping, or input to or training of AI) is prohibited under Article 8 of the Terms of Service. If you need data for research or analysis, please contact us.