Abstract

We introduce a novel method for using reflflectance to identify materials. Reflflectance offers a unique signature of the material but is challenging to measure and use for recognizing materials due to its high-dimensionality. In this work, one-shot reflflectance of a material surface which we refer to as a reflflectance disk is capturing using a unique optical camera. The pixel coordinates of these reflflectance disks correspond to the surface viewing angles. The re- flflectance has class-specifific stucture and angular gradients computed in this reflflectance space reveal the material class. These reflflectance disks encode discriminative information for effificient and accurate material recognition. We introduce a framework called reflflectance hashing that models the reflflectance disks with dictionary learning and binary hashing. We demonstrate the effectiveness of reflflectance hashing for material recognition with a number of realworld materials.