Regular biography
Naoto Yokoya is a professor at the University of Tokyo, affiliated with the Department of Computer Science. His research interests include image analysis and geoinformatics, with a focus on developing intelligent systems that understand and reason about the physical and semantic structure of the real world from visual and multimodal observations. His work spans computer vision, machine learning, and data fusion, with applications in Earth observation and remote sensing. Yokoya is also a team director at RIKEN and leads the Geoinformatics Team at the RIKEN Center for Advanced Intelligence Project (AIP).
Scholar-generated biography
Naoto Yokoya is a researcher at The University of Tokyo and RIKEN, specializing in Remote Sensing, Computer Vision, Machine Learning, and Data Fusion. His work focuses on advancing deep learning techniques for remote-sensing imagery classification, hyperspectral and multispectral data fusion, and hyperspectral image restoration. Yokoya has developed models such as SpectralGPT and ChangeMamba, which leverage spatiotemporal state space models for remote sensing change detection. His research also addresses challenges in hyperspectral unmixing, including spectral variability and endmember-guided learning. Yokoya's contributions span both theoretical and applied aspects of remote sensing, with a strong emphasis on interpretable and efficient AI solutions for environmental monitoring and urban land use analysis.