Mersive is in the business of making beautiful high-fidelity displays.  In this business, it turns out that what is beautiful may be, in part, determined by your customer’s sex.

Over the past six years, we’ve been developing software that transforms a set of ordinary projectors into seamless displays composed of tens of millions of pixels.  As our engineering and science teams work on new algorithms with the objective of a perfect seamless image, I often remind them that image quality, though sometimes characterized in objective measures, is really only important at a subjective level.  In fact, aspects of how the human visual system detects and responds to display anomalies can help drive the choice of an appropriate display algorithm – See “A Camera-Based Perceptual Smoothness Framework for Multi-Projector Displays.

As we work to improve our multi-projector color matching algorithms, it is important to take into account the importance of color surround on how a human observer will decide if two projectors match.   The impact of the observer on things like “how similar are these two flat panels in color temperature” is often overlooked and shouldn’t be underestimated.

One of the most compelling examples of subjectivity in visualization was demonstrated by a set of experiments at CUNY’s Brooklyn College that focused on the differences in men and women’s ability to see and process images. Israel Abramov, a researcher in the field of neuropsychology, demonstrated that the two sexes have core visual capabilities that differ.

Vision - men vs. women

For example, when groups of men and women are presented with vertical bars of differencing contrast on a computer screen that flicker, men are able to distinguish the bars as they become closer together and less distinct from one another – in other words, the results show that men’s visual systems  are better at seeing moving objects at distance.

On the other hand, Abramov’s team has also shown women are better at measuring subtle color differences.  When illuminating different controlled colors onto frosted glass, women were better at distinguishing subtle differences of hue near the middle of the visual spectrum – that is, they were able to see different hues of yellow that were not detectable by men.  They also discovered that in order for the two groups to report seeing the same hue, the men had to be shown a slightly longer wavelength of light.  Objects experienced as orange by women may appear slightly yellowish to men.

These interesting results create a whole new set of exciting questions.

  • Was there an evolutionary model that gave rise to these differences? (I was once caught in an unwinnable discussion with a group of female colleagues when I speculated it could be that color sensitivity is valuable for gathering food.)
  • Where in the visual processing pipeline do these differences arise?
  • Are they primarily cognitive in nature or are the differences partly embedded into the sensor apparatus itself?

I find the work relevant to what we do at Mersive for color-match projectors.  We modify projector wavelengths ever so subtly by operating beneath the perceptible levels of the average user to avoid perceived contrast loss. I suppose we may need to know our audience better!  Research like this may even be important to our AV integrator and installer partners.

Next time you install a display and the customer complains that it looks a shade too yellow, tell him to ask his female boss for a second opinion.

Share
About Christopher Jaynes

Jaynes received his doctoral degree at the University of Massachusetts, Amherst where he worked on camera calibration and aerial image interpretation technologies now in use by the federal government. Jaynes received his BS degree with honors from the School of Computer Science at the University of Utah. In 2004, he founded Mersive and today serves as the company's Chief Technology Officer. Prior to Mersive, Jaynes founded the Metaverse Lab at the University of Kentucky, recognized as one of the leading laboratories for computer vision and interactive media and dedicated to research related to video surveillance, human-computer interaction, and display technologies.

1 Comment for this entry

  • Johnk551
    September 5th, 2014

    This website was how do you say it? Relevant!! cgecceaekadd

Submit Comment