LAB LOGA page on LabLog, a review site (not the lab's official page)
kawa-lab (Kawamoto-Kera Lab)
Machine-translated · View original
The lab conducts research on computer vision, machine learning, and deep learning.
AboutAbout the lab
Research on AI that understands images and video.
The Kawamoto-Kera Lab studies computer vision, machine learning, and deep learning.
The lab works on techniques that show in a human-understandable form what an AI based its decision on, and on methods for efficiently reusing trained models.
It also studies the robustness of reinforcement learning used for robot control, and the understanding of human actions from video.
The lab belongs to the Information Engineering Course in the Faculty of Engineering, the Graduate School of Science and Engineering, and the Faculty and Graduate School of Information and Data Science, established in fiscal 2024.
Key terms
- Computer vision
- A field of technology that enables computers to read the objects and movements in camera images and video
- Deep learning
- A machine learning method that uses neural networks with many stacked layers to learn features automatically from large amounts of data
- Reinforcement learning
- A machine learning method that learns how to choose effective actions based on rewards obtained through trial and error
NewsLab news
Paper accepted at NeurIPS 2026
Paper published in Applied Intelligence
Paper posted on arXiv
Student wins Student Encouragement Award and Audience Award at MIRU2026
PhD student's paper posted on arXiv
ResearchResearch themes
The lab studies computer vision, machine learning, and deep learning. It works on explaining AI decisions, reusing trained models, reinforcement learning for control, and recognition of human actions.
*Illustrative imageExplainable AI
The grounds for a deep learning model's decisions tend to remain a black box. The lab aims to develop explanation methods that are theoretically well defined and reproducible, and that show the model's decision process in a form people can understand.
*Illustrative imageTransferring knowledge by adding and subtracting models
The lab studies a framework that extracts the knowledge in a trained model as a difference in parameters and manipulates it by addition and subtraction. This makes it possible to adapt to new tasks and integrate knowledge without large-scale additional training.
*Illustrative imageRobust reinforcement learning for control
In reinforcement learning used to control systems such as legged robots, the lab deliberately adds perturbations to action signals, body shape, and joint characteristics to probe the weaknesses of the learned controller. This serves as a diagnostic tool for building safe and reliable control systems.
*Illustrative imageAction recognition
This research aims to understand human actions from video and skeleton data. By integrating multiple types of data, such as skeleton data and first-person video, and by incorporating language information, the lab aims to make it possible to recognize actions the model has not been trained on.
MembersFaculty
The members of the Kawamoto-Kera Lab consist of faculty, project researchers, secretaries, collaborators, and doctoral, master's, and undergraduate students. Faculty from the Institute for Advanced Academic Research also take part in supervision.As of 2026
- ProfessorKazuhiko Kawamoto
- Associate ProfessorHiroshi Kera
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 kawa-lab (Kawamoto-Kera Lab)
Ishiyama LabFaculty of Informatics and Data Science, Department of Informatics and Data ScienceComputational methods and computer operation methods for hig…
Manabe-Yata LabFaculty of Informatics and Data Science, Department of Informatics and Data ScienceImage measurement, 3D data processing, computer vision, virt…
Kishimoto LabFaculty of Informatics and Data Science, Department of Informatics and Data ScienceSpecializing in cryptography, graph theory, flow networks, a…
Kubo LabFaculty of Informatics and Data Science, Department of Informatics and Data ScienceConstruction of new visual information technology connecting…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: “Artificial Intelligence (AI) and Robotics exhibition at the Heinz Nixdorf MuseumsForum” Sergei Magel/HNF / CC BY-SA 4.0 (cropped)
- Explainable AI: “PDC server room” Esquilo / CC BY-SA 3.0 (cropped)
- Transferring knowledge by adding and subtracting models: “Simplified neural network example” Mikael Häggström, M.D. Author info - Reusing images- Conflicts of interest: None Mikael Häggström, M.D. / CC0 (cropped)
- Robust reinforcement learning for control: “Anymal-robot-inspection-power-grid” Cvoellmy / CC BY-SA 4.0 (cropped)
- Action recognition: “Pole with surveillance camera and kill switch” Syced / CC0 (cropped)
- Related lab (Ishiyama Lab): “NCCS Discover Supercomputer (GSFC )” NASA Goddard Space Flight Center / NASA/GSFC/Bill Hrybyk / Public domain (cropped)
- Related lab (Manabe-Yata Lab): “VIUscan handheld 3D scanner in use” Creative Tools from Halmdstad, Sweden / CC BY 2.0 (cropped)
- Related lab (Kishimoto Lab): “Enigma (crittografia) - Museo scienza e tecnologia Milano” Alessandro Nassiri / CC BY-SA 4.0 (cropped)
- Related lab (Kubo Lab): “Rendering techniques example, path tracing, realistic, Blender Cycles” KaiaVintr / CC BY-SA 4.0 (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.