Affiliate Disclosure
If you buy through our links, we may get a commission. Read our ethics policy.

Pixelmator Pro image tool with CoreML, Metal 2 enhancements coming in the fall to iOS, macOS

Last updated

The Pixelmator developers have turned to Apple's CoreML machine learning toolset, and are releasing Pixelmator Pro in the fall to directly combat Photoshop's market dominance.

Announced on Tuesday, The new Pixelmator Pro has a redesigned interface and look. The older version's floating tool palettes are gone, in favor of expanded sidebar selections, and a new tab feature allows to switch between versions of the active image as work progresses.

A series of new tools allows the user to design layouts, and create vector graphics like Adobe Illustrator as well. Retouching images, and digital painting remain at the core of the app, and have been improved with the new version.

Pixelmator Pro retains nondestructive image editing, like its predecessor. A new Recipes feature allows users to save developed effects, use them in other images, and share them with other users.

Also added are custom key layouts for the MacBook Pro's Touch Bar, iCloud sync between iOS and Mac, and split-screen multitasking on the iPad.

Apple's CoreML has been utilized for a repair tool to remove and replace aspects of a photo. CoreML also facilitates a new quick selection tool, label layering based on content, and automatic image straightening based on horizon detection.

Pixelmator will still be maintained after Pixelmator Pro's release, according to the company. A specific release date is not yet known, nor is pricing.



13 Comments

stuke 15 Years · 123 comments

I love Pixelmator, have since Nenhance disappeared. But, I worry about loss of focus on P in favor of P Pro, and whether or not P Pro will grab more share against Affinity Photo and MacPhun Luminar. Now, if they turnout P Pro to be JUST LIKE Aperture with P's graphic prowess (meaning + a kick-butt DAM), Im sold!  I feel it's taking way too long for Affinty to get there and I'm not sure MacPhun is any closer. 

(Meanwhile, on an iMac late 2012, Aperture is still working after MacOS Sierra upgrade.)

frank777 22 Years · 5762 comments

It took them a full decade to make a Pro version, and I still don't see 'CMYK' anywhere on the feature list.

Maybe I missed it?

ylon 17 Years · 50 comments

Would love to see this compete with Adobe Fireworks and Sketch. Sketch just doesn't have the motivation to really do what Fireworks used to do in all respects and we need a REAL Fireworks killer since Adobe is such a loser and abandoned their main killer tool that no one seemed to really recognize due to their misinformation - disinformation campaigns.

Marvin 18 Years · 15355 comments

ireland said:
That icon looks hella familiar.

It looks like they were aiming for similarity to Apple's Photos app icon so that people would see Photos as where photos are kept and Pixelmator as how those photos would be edited:



The CoreML features will be really useful once the machine learning algorithms are able to detect, mask and remove complex objects accurately. This is an example of Apple's work on AI:

https://developer.apple.com/documentation/vision/classifying_images_with_vision_and_core_ml

Instead of Not Hotdog, they went with Is Banana:



They use a convolutional neural network for image analysis and they have an offline data set with 1000 object categories:

https://developer.apple.com/machine-learning/

This type of image detection is used for self-driving vehicles so they know what objects are around them:

https://hackernoon.com/creating-insanely-fast-image-classifiers-with-mobilenet-in-tensorflow-f030ce0a2991
http://selfdrivingcars.mit.edu/deeptesla/

One of the simplest benefits is the auto-layer naming in image editing but it can make very advanced tools for editing images especially coupled with Siri. You would be able to ask Siri to remove or move an object in a photo and it would be able to figure out where it was and do all the actions on it. If depth sensor data is added in to iPhone photos, doing perspective correct adjustments and surface editing (e.g shadow projections) will be even easier. Apple's portrait mode is just a basic example of computational photography, once all of these components come together, there's a lot more advanced and accurate things that can be done.