Naoto Yokoya
Naoto Yokoya works on machine learning for Earth observation. His research uses several kinds of remote-sensing data to study land and environmental change. Yokoya coauthored an ECCV study of self-supervised cross-modal super-resolution. The work connects information from different sensing modalities to image reconstruction. Yokoya was a coauthor of OpenEarthMap, a benchmark dataset for global high-resolution land-cover mapping. It appeared at WACV in 2023.
Timeline
-
2022
Learning across sensor types
Yokoya coauthored an ECCV study of self-supervised cross-modal super-resolution. The work connects information from different sensing modalities to image reconstruction.
-
2023
OpenEarthMap
Yokoya was a coauthor of OpenEarthMap, a benchmark dataset for global high-resolution land-cover mapping. It appeared at WACV in 2023.
-
2024
Land-cover change analysis
With colleagues, Yokoya coauthored ObjFormer, which studies land-cover changes using paired OpenStreetMap data and optical imagery. This is one concrete example of his group’s data-fusion program.
-
As of October 2026
Geoinformatics Team
Yokoya directs RIKEN’s Geoinformatics Team. The team applies image processing and machine learning to multimodal data for disaster and environmental monitoring.
Sources and credits
Photo: www.riken.jp; creator not stated. License: Not stated by source. Source: riken.jp. Cropped to a square and resized. Ask us to remove or replace this photo.