From www.pcmag.com
Dall-E 3 users can now edit AI-generated images rather than having to generate a new one from scratch to make changes.
It’s available now on desktop and mobile for those with a ChatGPT Plus subscription ($20/month), Enterprise or Team account, and developers through the API.
Just as the images are created from a text prompt, so are the edits. You can select the area that needs tweaking and then write a new prompt for that area, or you can type edits that apply to the entire image and Dall-E 3 will incorporate them.
(Credit: OpenAI)
When targeting a specific area, OpenAI recommends “selecting a large space around the area you intend to edit to obtain better results.” You can remove specific objects or swap them for new ones, as in the example below where Dall-E 3 changed the background to cherry blossoms. It can also adjust the facial expression to “happy,” “sad,” etc.
Don’t like what you did? You can undo/re-do sections as much as you like.
Recommended by Our Editors
(Credit: OpenAI)
The capability is likely one of the most significant upgrades to the image generator since OpenAI debuted the third version (“Dall-E 3”) of it in September. However, it’s also playing catch-up with its one of main competitors, Midjourney, which added the ability to edit images in August 2023.
Get Our Best Stories!
Sign up for What’s New Now to get our top stories delivered to your inbox every morning.
This newsletter may contain advertising, deals, or affiliate links. Subscribing to a newsletter indicates your consent to our Terms of Use and Privacy Policy. You may unsubscribe from the newsletters at any time.
[ For more curated Computing news, check out the main news page here]
The post OpenAI’s Dall-E 3 Image Generator Just Got More Precise first appeared on www.pcmag.com
