Intelligent Information Media Lab
Research overview
Our research themes conver a wide range of basic technologies and
applications for intelligent information-media systems that support
people in daily life.
As the basic technologies, computer vision, image understanding, and
other multi-media sensing and recognition techniques are widely
studied.
In addition, machine learning including deep neural networks for the
big multi-media analysis is also our focus.
These basic techniques are employed for various real-world systems for
human sensing, human modeling, and human support.
Main achievements:
- Two papers accepted to CVPR2024.
- Four papers accepted to IJCNN2024.
- One paper accepted to IEEE Transactions on Instrumentation and Measurement.
- One paper accepted to IEEE Access.
- One paper accepted to WACV2024.
- Two papers accepted to ICCV2023.
- One paper accepted to ICANN2023.
- IEEE Robotics and Automation Society Japan Joint Chapter Young Award given to ICRA2023 paper.
- One paper accepted to Neurocomputing.
- Three papers accepted to MVA2023.
- IEEE Nagoya Section Conference Presentation Awards 2023 given to CVPR2022 paper.
- One paper accepted to ICRA2023.
- One paper accepted to Japanese Journal of Applied Physics.
- PRMU Monthly Best Presentation Award given to our PRMU paper.
- One paper accepted to CVPR2022 Workshop(NTIRE2022).
- One paper accepted to CVPR2022.
- One paper accepted to IEEE Transactions on Intelligent Transportation Systems.
- One paper accepted to IEEE Transactions on Medical Imaging.
- Best Practical Paper Award at MVA2021 given to our MVA2021 paper.
- Six papers accepted to MVA2021.
-
Image enhancement such as Super resolution:
Images taken with a camera include various defects such as distant objects appearing small (difficult to recognize), blurring when the camera or subject moves, and objects that you do not want to see. These deficiencies interfere both when a person sees an image and when a machine sees the image and recognizes something. We study image enhancement technology to solve these problems.
-
Human activity recognition with big data learning:
For human pose, motion, and activity recognition, we utilize not only training data given by manual annotations but also unsupervised data (e.g., images, videos, and texts in the Internet) in order to improve the performance.
-
Human group and crowd analysis:
Activities of human group and crowd are analyzed for recognizing highly-contextual events in the real world.
-
Human motion sensing and estimation:
Natural motions of a human body are measured by optical sensors (e.g., cameras). These accurately-measured motions are also modeled by machine learning in order to explore individual characteristics and skills. Human motions are also observed by cameras and estimated in observed images and videos. These estimated motions are useful for monitoring, HCI, HRI, and other applications.
-
Human-like Motion Generation and Prediction:
Human and robot motions are generated and predicted for a variety of applications such as human motion animation, human trajectory prediction, and robot motion planning.
-
Video recognition for inside and outside vehicles:
For smart vehicle control, we develop activity and condition recognition methods for passengers and pedestrians.
- For more detailed information, see here.
Professor
Researcher
D
M2
M1
Kyotaro Tokoro
Kazuki Matsumoto
Yota Urano
Ryuki Tezuka
B4
Takuya Seno
Takato Moriki
Kaname Yokoyama
kento Kawai
Kohei Kuribayashi
Shozaburo Hirano
