Skip to main content
. Author manuscript; available in PMC: 2014 Dec 1.
Published in final edited form as: IEEE Trans Image Process. 2013 Dec;22(12):5146–5157. doi: 10.1109/TIP.2013.2282120

Figure 5.

Figure 5

Relative preferences by group (Sharp and Smooth) for high and low ratings of video-content categories (a) Faces, (b) Human figures, (c) Nature, and (d) Man-made Objects. The Face video-content category showed a response pattern that differed from the other three video-content categories, with enhancement being less preferred for high-rating (Face) video clips than for low rating (Non-Face) video clips (note the reversal of order of filled and open symbols from top to bottom). Error bars show 95% confidence intervals. To improve clarity, the group symbols are plotted with small offsets.