How to colorize your old black and white photos
Life in Technicolor is just better.
You might be worrying about droids powered by artificial intelligence taking over the planet in the long term. But in the meantime, you can use that same technology for all kinds of helpful tasks. Colorizing old black and white photos is one of them.
Several platforms can give new life to your old snaps with the help of neural networks. These intricate pieces of code use cues in the shading of black and white photographs to identify a grayscale value, and then cross that information with vast databases of sample pictures to find its colored match. This allows them to identify a specific hue and add it back to the image. This all happens in the background, though—the only thing you need to do is click on the right place.
But before you get excited about reviving all of your old family albums, you should know this isn’t an exact science. Because black and white photos don’t actually have any color information in them, these apps are using machine learning to make educated guesses about the tones in an image. The quality of those guesses, and the accuracy of the photo, will depend on the platform you use, so results may vary between each one.
Few colorization tools are as straightforward to use as the @colorize_bot account on Twitter. You simply tweet your black and white photo to it, and you get a full color version in return. It really is that simple.
If you want to get an idea as to what kind of results you could get, take a look at the account’s Twitter feed before you submit your own picture. There is a processing limit of one photo per user per hour to manage demand, but once you send out that tweet, you should get a reply within 10 minutes.
The bot was put together by Twitter user and computer science student @asdrubalzp, and is powered by the colorization engine at Hotpot AI. You can also visit the Hotpot AI site directly to upload your pictures and get them colorized.
[Related: Edit gorgeous photos right on your phone]
Although you can also work on files you’ve already scanned and uploaded, the app opens up your camera automatically. Once there, snap a photo of a physical black and white image, and press and hold the shutter button to see the result in seconds. The app also crops the pic automatically for you, and you can manage any images you take or import in a colorized album inside the app.
A free Colorize account will let you scan, process and store a bunch of your pictures, and though the actual number is not clear, it is not unlimited. If you don’t want to find out what that cap is, you’ll need a subscription to use Colorize, which will set you back $6 a month or $30 for two years. The good news is that you’ll have a lot of opportunities to test the platform before you decide if the app is worth your money.
Ancestry services have an interest in helping users restore old family photographs. MyHeritage is no exception, offering users a free online app for converting black and white images to color very easily.
On your browser, go to the website and click the Upload photo button or simply drag and drop your picture to start the upload and conversion process. You don’t have to pay anything to use the colorization tool, but you will have to sign up for a free MyHeritage account to see and save the end results.
MyHeritage also has a color boost tool for faded photographs, which is worth trying on pictures that have lost a lot of their luster.
This platform offers a broad suite of deep learning tools for businesses to use, but it also provides a free online image colorizer as a demonstration of what its artificial intelligence systems can do. And the results are rather impressive.
On the web app you can paste an image URL, upload a photo from your computer, or even try out one of the example images to see how it all works. A result appears in seconds, and you can play around with the on-screen slider to see the difference between the original picture and the processed version.
There’s nothing in the way of image management here, but the tool is totally free and you can download the new colorized version or even a half-and-half comparison picture to your computer. The only restriction is that your finished images will have a small watermark on them.
This is another simple colorization app and comes as one part of a suite of tools covering all aspects of photo restoration. Besides the web platform, there are apps for Windows, macOS, Android, and iOS, so it’s one of the most comprehensive options available to you.
Your browser is a good place to start because you won’t need to register for an account. Just go to the website, drag and drop a black and white image into the tab, and Colorizer will do the rest. Alternatively, click on the photo icon to choose a file from your hard drive.
There are limitations on the free use of Colorizer which vary depending on which of the apps you use. You might only be able to upload images up to a certain size, or the app might crop the final result, or stamp a watermark on it. Removing all restrictions requires a subscription, starting at $6 a month.
[Related: How to shoot great Instagram photos]
Adobe Photoshop CC
You’ll have to pay at least $10 a month for Adobe Photoshop CC for Windows and macOS , so it’s unlikely that you’ll choose this tool over the other alternatives on this list if all you want is to pump some color into your pics. However, if you already have Photoshop, it’s worth knowing that this functionality now comes built in.
To find it, go to Filter and Neural Filters. Then click the beta filters button (it looks like a classic lab beaker), click on Colorize, and turn on the toggle switch to start applying the effect. Keep in mind that, at the time of writing, this feature has a ‘beta’ label on it, which means it’s not fully finished.
The advantage of using Photoshop for the job is that you can tweak the filter and adjust spots of color if needed. This comes in handy if you know what color a particular object or background is, as you can add in this information and adjust the filter accordingly.