Department of Advanced Information Technology, Faculty of Information Science and Electrical Engineering
Department of Advanced Information Technology, Graduate School of Information Science and Electrical Engineering
Department of Electrical Engineering and Computer Science, School of Engineering
In the Laboratory for Image and Media Understanding (LIMU), our goal is to establish a novel framework to (1) retrieve social information from observation data obtained with various sensors and (2) to create innovative content for the society by analyzing those data. While developing the tools necessary to build such a framework, we carefully design the algorithms so that anybody in the society can later interact with the cyber-physical world to improve analysis performances and user experience. To this end, the research in the lab is focused on three main axes: video processing, agricultural Internet of Things, and learning analysis. The goal of video processing is to make the computer “understand” the world. This means to develop techniques to analyze, recognize and process visual information such as images and videos. In the lab, we develop fundamental methods such as object detection and anomaly detection using machine learning for efficient scene understanding. In this research, we are making much effort to establish the next-generation agricultural Internet of Things. Various sensors are installed in farms and captured data are analyzed using most recent machine learning technologies. In the lab, captured visual data are used for application in the agricultural world. Results of our analysis on plant growth, vegetable production, and harvesting are provided to our farmer collaborators with various information that helps for everyday support and production improvement. On the other hand, we also analyze educational big data such as students’ learning activities collected from digital textbook systems and learning management systems. The various educational data are analyzed to provide real-time feedback systems for visualizing student’s learning activities and teaching materials recommendation systems personalized for students individually. These results can be used to develop services leading to a more efficient and sophisticated society.