AI-Powered Software Makes It Incredibly Easy to Colorize Black and White Photos

Original photo: John Rooney/AP; color modification: Richard Zhang

AI-powered software that can automatically colorize old black and white photos exists, but it’s often far from perfect. In comparison, manually colorizing an image in Photoshop yields stunning results, if you’ve got lots of time and impressive skills. But a new app, developed at the University of California at Berkeley, cleverly merges both approaches so it’s easy to accurately colorize a black and white pic.


A year ago, Richard Zhang and his team at the University of California at Berkeley demonstrated a neural-network powered app that could automatically add color to black and white photos. After studying millions of images for reference, the app had learned how various features in color photos were tinted, and could apply what it learned to images it had never seen before. The results were often impressive, even if the app got all the colors wrong.

To deal with those mistakes, Zhang and his team have now developed a follow-up app that uses the same highly-trained convolutional neural network to automate the process of adding color to a black and white photos, but with the addition of user-added cues and hints that help the software, and the underlying algorithms powering it, produce more realistic results.

The app starts by taking its best shot at colorizing a black and white source image it’s been fed, and in the process it creates a small palette of suggested colors. The user can then refine the colorization process by adding simple color markers to the source image. For example, they can specify a plant is a certain shade of green, or choose a specific skintone color for a person, just by placing tinted dots right on the source image in the app.

Original photo: Dorothea Lange; color modification: Richard Zhang
Original photo: Dorothea Lange; color modification: Richard Zhang

The neural network continues to work away in the background, updating the colorized results every time a new marker is added by a user. So instead of having to spend hours in Photoshop painstakingly painting in hundreds of different shades, the colorization process happens in just minutes. It’s intelligent enough to limit where colors are applied, so a handful of dots are all that’s needed to bring an image to life.


Want to try it for yourself? There’s no word on when or if Adobe plans to include capabilities like this directly in Photoshop, but Zhang has made his Interactive Deep Colorization app available for download on GitHub if you want to take a stab at colorizing some cherished photos of your great grandparents.

Update: The photo captions in this article have been updated to credit both the research team that colorized the photos as well as the photographers of the original images.



The Noble Renard

The last picture is a good example of why what this software can produce may look good, but there’s no evidence that it’s accurate. The choice of a yellow blanket for the baby and a pink shirt for the girl were entirely the decision of the person creating it. In reality they could have been completely different colors.

This is why colorizing black and white photos is a neat trick but generally rarely improves the image from an artistic perspective. It’s a less accurate reflection of the scene than the black and white image.