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. 3. Visual recognition as a function of spatial scale.

Fig. 3

(a) Extreme close-up views, (b) close-up views, (c) regular views, and (d) zoomed-out views of (top row) a car and (bottom row) a table. Visual recognition is driven by (a) 2-D and 3-D textures properties, (b) material properties, (c) 3-D shape and object properties, and (d) scene properties. In this work, we will use images from the Flickr Materials Database (FMD) [Sharan et al., 2009] that depict spatial scales in the range (b)-(c).