Apple Intelligence: Image Playground

Apple has gotten into what society is calling "Artificial Intelligence", or "AI", except they have put their own spin on the term. One of Apple’s core tenets is protecting user privacy, and they attempt to do so whenever possible. In order to accomplish this, Apple attempts to protect user privacy through a combination of performing as many requests on the device as possible. However, for advanced requests that may not always be possible. For those requests there is a feature called Private Cloud Compute. Private Cloud Compute is a set of servers run by Apple that are capable of handling your request, providing the result, and then the server is completely erased. If you need more details, be sure to check out the post that is the Introduction to Apple Intelligence.

Apple is releasing the Apple Intelligence features in batches. The first set of features released were done so in iOS 18.1, iPadOS 18.1, and macOS Sequoia 15.1. This set included "Typing with Siri", "Summarization and Mail", and "Writing Tools". The second set of Apple Intelligence features are being released with iOS 18.2, iPadOS 18.2, and macOS Sequoia 15.2. There are four new features for this batch. This article will cover a feature called "Image Playground".

Everyone has a different set of talents. Some people can sing, others can entertain through comedy, yet others can write, and some can even draw. There are many who wish they could create works of art, and yet they cannot.

When Apple introduced Apple Intelligence, one of the features promoted was the ability to generate images. The way that this would be done is through a dedicated application called Image Playground. Sometimes you have an idea of the type of image that you want, but you have no idea how to get started. Image Playgrounds can help.

Image Playground is designed to allow you to generate an image based upon prompts that you give it. For example, you could describe something like "Show an Alien farming with a sci-fi theme", or you could say "Corgi and a goat, kayak on a lake", or even "Cow wearing a blue party hat", or just about anything else you can think of, and it will be generated.

Generating Images

The way that you generate images is quite simple; you can use one of two approaches. The first is to add individual words to help describe what you are looking for. The second option is to use full sentences to indicate what you are looking for.

Once you have added your words or description, an image should be generated. You can add additional words to help refine the image that has been generated. In fact, there will be four variations generated. The images that are generated are square and come in at 1024 pixels on each side. You can cycle through the four generated images, and if you do not see one that is what you are expecting, you can click on the right arrow, and it will generate another new image using the same phrase and words that you used previously.

After an image is generated, you may want to go back and change the description that you have chosen. This can be done by tapping or clicking outside of the image. When you do this, you will be brought back to the "Description" screen with all of your selections or entered words.

Tip

A tip for when you generate an image is to be as descriptive as possible. The more detailed, the better the results will be. Each prompt item can be a maximum of 100 characters, but you can add additional phrases.

Suggestions

Sometimes you may not be able to think of something to generate. For those instances, there are the "Suggestions". Suggestions can be one of a variety of types. These include:

• Themes
• Costumes
• Accessories
• Places

For each of these types, there will be twelve options. You can click on any of them and add them to your image.

Using People and Pets

It is possible to generate images that do not include people; however, once you add a word that could even possibly relate to a person, Image Playground will REQUIRE you to add a person. This does not necessarily need to be a photo of a person, although that is the intention. Instead, you can use an "appearance".

Appearances

An "Appearance" is one of five skin tones matched with one of three appearances of the type of person. One thing that did not make sense was why there were three different options.

Using Photos

If you prefer to create an image based on a photo, you can do so. You can use any photo that you would like, and it will generate an image from the photo. The photo could be of a person, a pet, a nature scene, or any other photo. Image Playground will do its best to try and generate an image to your specifications.

Styles

When Image Playground was initially introduced, it was indicated that there would be three different styles: Animation, Illustration, and Sketch. This last one, Sketch, is not currently present, and it is unknown if it will be added. However, you can still generate an Animation or Illustration. Sketch may be added in the future, but for the time being, it is not there.

If you have generated an image in one of the available styles, you will be able to switch the style by clicking on the "Style" button and then clicking on another style. The image will be regenerated in the requested style. You might think that if you switch back, the image would be regenerated again, but it will not. Instead, the previously generated image will be shown.

Saving Images

Once you have finalized the image, you have a few options. The first is that you can save it. When you do this, it will appear in your Image Playground library. In the instance that you have edited an image and if you attempt to save, you will be prompted whether or not you want to overwrite the image or save as a duplicate. The popup will appear like this:

One thing I would recommend is to save as a duplicate; you can always delete an image later if desired.

Sharing Images

If you want to show the generated image to someone, you can also share an item. To do this, you can use these steps:

1. Locate the image you want to share.
2. Click on the "Up Arrow" button to bring up the system Share Sheet.

Alternatively, you can tap and hold, or right-mouse click, on the image and then select the "Share" item to open the Share Sheet.

Regardless of the manner in which you bring up the Share Sheet, you can then perform any action defined within your Share Sheet, including copying, saving, sending, emailing, or even AirDropping the image.

Image Playground within Notes

One of the features of the Notes app is the ability to draw directly within the app. This feature was introduced with iOS 9 in 2015 and was designed for the original iPad Pro, which introduced the Apple Pencil. Since its introduction, the Notes app has gained even more features, including the tool palette. With iPadOS 18.2, there is a new tool called "Image Wand".

The Image Playground tool is integrated directly into Notes and can be used to take a drawn image and generate a photo from the drawing. To generate an image from a drawing, perform the following steps:

1. Select the drawing that you want to use as a basis for the generated image.
2. Tap and hold to bring up the menu.
3. Tap on "Add to Playground" to open up the image generation popup.
4. Enter some text to help describe what you are looking for.
5. Hit the "up arrow" to generate the image.

Just like Image Playground, four different options will be shown. You can swipe between them to find the right image. Once you have found the proper image, tap on "Done" to add it to the current note.

Generating from Text

Being able to generate an image from a drawing is useful, but sometimes you want to augment a piece of text with an image. If you have a description that you want to use to generate an image, perform the following steps:

1. Select the text you want to use to generate an image.
2. Tap and hold to bring up the menu.
3. Tap on "Add to Playground" to bring up the image generation popup.
4. Swipe between the generated options until you find a suitable image.
5. Tap on "Done" to have the picture added to the current Notes document.

I tested using this text: "Imagine, if you will, a dilapidated house with broken windows, shudders hanging off of the sides, on a hill". This was one of the results:

Gallery of Examples

Here is a gallery of examples from Image Playground:

Photo-based

Photo-based

Closing Thoughts

Image Playground is capable of taking your input, whether it be text or an image, and generating an image. Whether you use text only or ad an image, you can add additional words or description to create the image that you are looking for. It may take a few tries to get the image you are looking for. One tip is to be as descriptive as possible.

For your images, you have a couple of different options for styles, including Animation and Illustration, and the image will be 1024 x 1024. After you have generated your image, you can share it with others, including emailing, AirDropping, or even messaging, depending on what you need.

It is hoped that Apple increases the physical size of the images and maybe even allows different sizes because it would work better in some situations than others.

Be sure to check out all of the other articles in the Apple Intelligence series.