Mihoko Shimano

Papers

Total Citations

H-Index

About

Mihoko Shimano is a computer vision researcher whose work focuses on the intersection of material recognition and spectral imaging. Her primary research area involves the computational analysis of surface properties—specifically, how to estimate wetness and color from a single multispectral image. This work has significant implications for autonomous systems, robotics, and consumer applications, enabling machines to detect slippery roads, muddy trails, or the freshness of produce. Shimano’s major contributions include pioneering methods for recognizing wet surfaces and quantifying their degree of wetness, a challenging problem because water alters both the color and reflectance of materials. Her 2017 paper, "Wetness and Color from a Single Multispectral Image" (9 citations), and its 2019 follow-up, "Estimation of Wetness and Color from a Single Multispectral Image" (6 citations), demonstrate how multispectral data can disentangle the effects of water from intrinsic surface color. By showing that wet surfaces darken in predictable ways, she provided a foundation for robust visual perception in real-world environments. Though her citation counts are modest, Shimano’s work is notable for its practical relevance and interdisciplinary appeal, bridging computer vision, optics, and materials science. Her research has been cited in studies on autonomous navigation, agricultural inspection, and robotic manipulation, highlighting its potential to enhance machine understanding of the physical world.