There’s nothing quite like the excitement of new AI: the creation, the failures, and the weird, uncanny valley images, and it’s been no different with the Samsung Galaxy Z Fold 6 (unveiled at Galaxy Unpacked in Paris) and its remarkable Galaxy AI capabilities, which are on full display in Sketch to Image in the Photos app.
I’ve written about Sketch to Image in Notes before and its capabilities for creating something from nothing, but I hadn’t really dug into Galaxy AI’s photo manipulation capabilities, which are awesome.
To test them, I took a series of selfies in New York City’s Times Square. The population density and sheer craziness provided what I thought would be a particularly challenging environment for the cloud-based generative image engine.
Picture 1 of 5
As it turned out, the real challenge was figuring out how to draw in a way that the Galaxy AI on the Samsung Galaxy Z Fold 6 could understand and then interpret in an artistic, or more often photorealistic, way.
Much like the Sketch to Picture feature in Notes, very detailed sketches didn’t work and at some point the system asked me to draw objects smaller or closer together. A bunch of disconnected lines usually doesn’t work.
However, Galaxy AI can infer a lot from just a few lines.
The more you play with Sketch to Image, the more you’ll want to use it. I ended up creating about a dozen different things to sit next to me on a stone bench.
I start by taking a photo, and for my images, I tried to leave a little space on either side of me because I knew I would generate companions, or at least I would try. Then I jump into the Samsung Galaxy Photos app on the Galaxy Z Fold 6, open an image, and select the little Galaxy AI icon. After that, I select the Sketch to Image button and start drawing. While I can change the color and width of the pen, this is not a drawing platform. It is difficult to create shapes that look like something, at least not without practice. Also, the system has an annoying habit of smoothing out lines and shapes, which you can undo by tapping the Undo icon.
Eventually, I figured out how to do minimal work for maximum output. A caterpillar on my sleeve started as a series of bumps with a few lines for legs and antennae. To generate a blue bird on my head, I made the line thinner but, more importantly, changed the pen color to blue.
Galaxy AI doesn’t just print or stitch these objects onto your photo. In all of my images, the photorealistic objects are properly lit for the scene and they cast shadows in space. It often looks like they’ve always been there.
The dog image was perhaps one of the more detailed sketches. Galaxy AI did a good job, but it seems to just look at the sketch, figure out what you want, and then create a photographic image that represents your intentions. There’s little connection between my sketch and this dog. While this dog looks pretty good, there are some weird AI things in it. One of his jowls has its own paw, and he has some very strange fur or hair near his belly. Otherwise, kudos, Galaxy AI.
Even though Galaxy AI does Sketch to Image in the cloud, the process is pretty quick, usually taking just a few seconds before I can choose from a few generative options. You can save a copy of what you like, but once you do, all the other options are gone forever.
All final images receive an “AI Generated Content” watermark that you cannot remove but can easily crop.
If the dog is my favorite image, the caterpillar comes in second because it is the most subtle. The beach ball was a nice effort. That rocket behind me looks almost nothing like my sketch.
At one point I tried to draw a head sitting next to me, but the best Sketch to Image could do was put a flat face on the bench.
You can add multiple fake objects to an image, but you’ll have to save a copy between each one. It might be best to do what I did and delete something you didn’t like, then add a new object that you wanted in the image. I did this with a family standing behind me in Times Square, deleted them, let Galaxy AI fill the space with whatever might be there, and then added the blue bird on my head just like that.
Sketch to Image for photos is a fun tool that lets you reimagine images and connect with new photographer friends like my white avatar friend. Could people take advantage and try to fake images? Sure. They could, crop out the watermark, and share it, but most of the Galaxy AI-generated stuff doesn’t quite look real, and I have a feeling a lot of these AI-generated objects will start to look the same. Soon, people will be able to recognize Galaxy AI-generated content at a glance. Or at least, I hope they will.