Skip to main content
. Author manuscript; available in PMC: 2014 Jul 1.
Published in final edited form as: Int J Comput Vis. 2013 Feb 19;103(3):348–371. doi: 10.1007/s11263-013-0609-0

Fig. 4. CURET vs. Flickr Materials Database.

Fig. 4

The projection of the first two principal components (PCs) of texton histograms are shown for all images in (left) the 61-class CURET database [Dana and Nayar, 1998] and (right) the 10-class Flickr Materials Database [Sharan et al., 2009]. The textons were derived from 5×5 pixel patches as described in [Varma and Zisserman, 2009]. The colors indicate the various texture/material categories. These plots demonstrate that CURET samples are more separable than Flickr.