Three Predictions on the Future of Color Correction

The moment you begin thinking about the future of color correction, there’s some clear good and bad news. On one hand, the sheer volume of motion images being captured and distributed across the globe is greater than ever before, and it’s growing exponentially. This would seem to promise that color grading has nowhere to go but up. That is until you realize that a large majority of this content will never be professionally mastered. It’s being produced too quickly, too cheaply, and with too little emphasis on visual polish for its creators to bother. It’s plain to see that well-graded content is rapidly losing market share.

But the future may not be as grim as it looks. As we’re going to discuss shortly, new tools for crafting polished images are right around the corner and will offer better, faster, and cheaper solutions to creators. Just as important, while professional color grading’s slice of the pie may be shrinking, the pie itself is growing even faster, yielding a net increase of creators looking for quality color work.

We don’t need to go out on a limb to observe that AI and machine learning will be the defining technologies of the next several decades — in fact, they may be the last things our species needs to invent.

Image mastering is sure to be a beneficiary of these technologies — in fact, it already is. More and more of the work traditionally requiring human operators — including correction, balancing, and quality control — will be handed over to software that will get the job done faster and better. As a result, colorists like me will be sought out not for our technical acumen, but for skills that rest at higher altitudes on Hans Moravec’s landscape of human competence.

• Presently, the most extensive and high-quality data we have about human visual preference in motion images comes from R&D which assumes a very low upper limit on the reproduction of color and luminance (corresponding with the physical limitations of projected motion picture film). Today’s displays, and even more so tomorrow’s, have far greater limits, so we have to figure out what pleases or repels the human eye within this larger canvas. Should we attempt to adapt the lessons of motion imaging’s first century, or would we get better results by starting anew?
• When our display has the ability to reproduce enough dynamic range to cause the iris to open or shut, how do we handle transitioning the eye between high and low exposures without creating an unpleasant experience?
• When a viewer is inside of, rather than looking at, an image, how can we use color, contrast, and focus to effectively lead or mislead their gaze to most effectively tell a story? What degree of active control over these attributes is desirable for the viewer? How much of these attributes should be “cooked” into the captured media, and how much should we be able to adjust in post-production? How can we harness technology to achieve this optimal degree of control?