After entering the watermark for DALL-E 3 images OpenAI has updated its chatbot Chat GPT with a really interesting function. As you know, through it it is possible to create images using the power of the proprietary software mentioned above. Well, the artificial intelligence giant now allows you to edit images created by DALL-E 3 quickly and easily without using external tools like Canva.
OpenAI just added image editing to DALL-E 3 - here's how to edit them
First you need to use ChatGPT 4 since it is the only version to integrate DALL-E 3. Using version 3.5 it will not be possible to create any images. Once logged in, simply enter a prompt in the bottom bar to generate an image. In my case, as you can see below, I wrote “Create an ultra-realistic portrait image of a woman wearing a cap in the middle of a Paris street"
Once DALL-E 3 has generated the image it will be enough instead sopra. This will allow us to go into specifics and open up a sort of menu new which you can see at the top right (photo below):
By clicking on the first icon on the left (“Select“), you will see a brush style selector which will allow you to select a specific section of the image. This will be the one that will be modified by entering a specific prompt. In the example below I selected the shoulder of the woman portrayed and entered as a prompt: "Add a parrot with an eye patch"
Now, clicking on the “i” (“Prompt“) to the right of the ChatGPT menu will show the prompt exact it's correct of the final modified image. This can be useful for generating or editing other images with DALL-E 3 using the same gen_idthat gen_id of an image generated by artificial intelligence is nothing other than the fundamental model, the image matrix.
Image editing works also on iOS and Android.
It is worth mentioning that the OpenAI GPT Store, that is, a real shop that allows you to monetize through i Custom GPTs.
