The Effect of Perceptual Optimization on Color Space Navigability

Trevor Dalton Canham, I. Scott MacKenzie, Richard F. Murray, Michael S. Brown

Keywords: Color spaces, color correction, color perception, psychophysics, user interfaces

TL;DR: An experiment was conducted to test the effect of decorrelating perceptual attributes on the ease of use of direct color manipulation tools.

Abstract: An experiment with 22 participants examined the effect of perceptual optimization on color space navigation in a direct manipulation task. Participants made ten color matches using a 2+1 degrees of freedom trackball to control saturation, value and hue in the HSV color space and chroma, intensity and hue in the HCI color space. (HCI is an IPT-based color space.) Participants completed the task significantly faster with the HSV space, at 41.3 s per color pair compared to 46.7 s with HCI. However, the difference in accuracy between color spaces was not significant. Correction path analysis suggests that the benefit of HSV's uniform geometry outweighed HCI's perceptual optimization. This was demonstrated by a significant difference in the number of times participants encountered gamut boundaries, which for HSV was about 0.5 times per trial versus 1.5 for HCI. Participant responses and comments revealed that a separated control for the hue attribute was particularly useful for navigating the color space.

Track: HCI/visualization

Summary Of Changes: Reduced white space on page 8, as suggested by paper chairs for the final camera ready version.