Keynote

Abstract:
Mediated Reality technology enables modifications to visual reality by encompassing both the addition of information (Augmented Reality; AR) and the subtraction of information (Diminished Reality; DR). Use cases range from re-furnishing homes to on-site urban planning by virtually removing existing objects and adding new ones as replacements. These reality modifications not only change our vision but also impact our perception of touch, weight, and task loads within the presented reality, thereby posing both technical challenges and providing deeper insights. In this talk, I will cover recent advances and future challenges in this research area

Speaker Bio:
Dr. Shohei Mori is a junior research group leader at the Visualization Research Center (VISUS), University of Stuttgart, Germany, and a Guest Associate Professor (Global) at Keio University, Japan. His research focuses on computational Mediated Reality, a fusion of Augmented Reality and Diminished Reality, with an emphasis on human-centered visual computing. His particular interests lie in core computer vision and graphics problems inherent in Mediated Reality. He has received multiple scientific awards, including Best Paper Awards (IEEE ISMAR 2023, IEEE VR 2022, IEEE ISMAR 2021, IEEE VR Workshop KELVAR 2020), Best Demo Award (IEEE ISMAR 2016), and Best Presentation Awards (KJMR 2015, KJMR 2014).

Tutorial

Abstract:
Point cloud is a data representation of 3D scene that is widely used in computer vision and robotics. Recently, deep learning methods have been applied to point cloud data for various tasks such as classification, object detection, segmentation, and completion. In this talk, I will introduce the basic concept of point cloud deep learning and its applications.

Speaker Bio:
Dr. Naoya Chiba is an associate professor, D3 Center, Osaka University. He received his Ph.D. in Information Science from Tohoku University in 2020. His research interest includes geometric deep learning, 3D computer vision, and robot vision. Especially, he has been working on point cloud deep learning for 3D scene understanding and reconstructions. In these days, he is focusing on application of the 3D deep learning methods in related research fields such as material informatics, computational mechanics, and virtual reality.