Skip to main content
. Author manuscript; available in PMC: 2015 Aug 26.
Published in final edited form as: Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2014 Jun;2014:2059–2066. doi: 10.1109/CVPR.2014.264

Figure 5.

Figure 5

(a) Performance of the epitomic dictionary model (without epitome position encoding) and the non-epitomic baseline for different epitome/patch sizes, as a function of dictionary size K. (b) Effect of encoding the epitome position at different detail levels. (c) Comparison of the epitomic model (with or without position encoding) and the non-epitomic baseline, for the same total histogram length.