Papers
9
Total Citations
65
H-Index
4
About
Ko Nishino is a leading figure in computer vision whose research spans computational surface modeling, material recognition, and egocentric scene understanding. His work bridges the gap between controlled laboratory reflectance measurements and real-world image-based representations, as exemplified by his highly cited 2020 paper on differential viewpoints for ground terrain material recognition (31 citations). Nishino has made pioneering contributions to understanding surface wetness from multispectral imagery, developing methods that enable autonomous vehicles to detect slippery roads and robots to navigate muddy terrain. His recent work on "view birdification" introduces novel learning-based approaches for recovering ground-plane crowd trajectories from single egocentric views, with applications in mobile robot navigation and localization. In 2025, his MAtCha Gaussians work achieved high-quality 3D surface mesh recovery and photorealistic novel view synthesis from sparse views. As guest editor for a special issue on RGB-D vision, Nishino has helped shape the field's direction. His research consistently addresses fundamental challenges in visual perception, from material properties to crowd dynamics, with clear practical implications for robotics, autonomous systems, and augmented reality.
Research Focus
Key Achievements
Top Papers
- 1Differential Viewpoints for Ground Terrain Material Recognition31 citations · 2020
- 2Wetness and Color from a Single Multispectral Image9 citations · 2017
- 3Estimation of Wetness and Color from a Single Multispectral Image6 citations · 2019
- 4
- 5
- 6
- 7
- 8
- 9