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. 15. Comparison to Varma-Zisserman.

Fig. 15

We ran Varma-Zisserman’s system [Varma and Zisserman, 2009], which uses a nearest neighbor classifier, on our feature sets. The training rate for a nearest neighbor classifier is always 100%, so we do not show the training rate in these plots. These results demonstrate the power of using our features vs. fixed-size grayscale patch features (23.8% accuracy) for FMD images.