High-resolution Hyperspectral Imaging via Matrix Factorization: In this work, we introduce a simple new technique for reconstructing a very high-resolution hyperspectral image from two readily obtained measurements: A lower-resolution hyperspectral image and a high-resolution RGB image. Our approach is divided into two stages: We first apply an unmixing algorithm to the hyperspectral input, to estimate a basis representing reflectance spectra. We then use this representation in conjunction with the RGB input to produce the desired result. (With Rei Kawakami, John Wright, Yu-Wing Tai, Yasuyuki Matsushita, and Katsushi Ikeuchi). [Project Page]

Multi-spectral Imaging: We present a novel active imaging approach that uses optimized wide band filtered illumination to obtain multi-spectral reflectance information. Our optimization algorithm utilizes light source and camera spectral information in order to maximize the signal strength and the robustness to noise. Through the use of active wide band illumination, our system can obtain material reflectance information in the presence of moderate (indoor) unknown ambient illumination. Our method is very simple and does not require special equipment. It can be used by photographers to obtain material properties in uncontrolled environment and to synthesize captured scenes under arbitrary illumination. (With Cui Chi and Hyunjin Yoo) [Project Page]