(a) Temporal interaction of light and material. A correlation image sensor correlates the reference signal with a mixture of direct (black) and indirect (green and red) components, as indicated in blue. (b) Visually similar but structurally distinct material samples in RGB. Top: paper; bottom: wax. (c) After depth normalization, the raw measurements of materials in (b) become separable.


We propose a material classification method using raw time-of-flight (ToF) measurements. ToF cameras capture the correlation between a reference signal and the temporal response of material to incident illumination. Such measurements encode unique signatures of the material, i.e. the degree of subsurface scattering inside a volume. Subsequently, it offers an orthogonal domain of feature representation compared to conventional spatial and angular reflectance-based approaches. We demonstrate the effectiveness, robustness, and efficiency of our method through experiments and comparisons of real-world materials.


Paper [MaterialToF_Su2016.pdf (10MB)]
Supplementary [MaterialToF_Supplementary_Su2016.pdf (3MB)]


Dataset [MaterialToF_Dataset_Su2016.zip (1.2MB)]


Code [MaterialToF_Code_Su2016.zip (xxMB)]

All images are © IEEE 2016, reproduced here by permission of IEEE for your personal use. Not for redistribution.