Photo editing used to be a field reserved for a few privileged people who mastered graphic design, but in recent years we have seen how artificial intelligence has advanced by leaps and bounds in the imaging field, especially opening it up to any user without requiring editing knowledge. Such is the case of DALL·E, an AI capable of generating content in a photograph thanks to its neural network.
The results are astonishing and in a matter of seconds we can remove a person or any element that bothers us in a snapshot, or even generate more background to expand the image’s canvas.
How to use DALL·E to erase elements from your photo
OpenAI, the company behind the DALL·E project, allows us to use the service for free – they give us 15 tokens to use as a trial – after registering on their website.
Once registered and logged in, simply go to their website and click upload an image to upload the photograph you want to edit.
When uploading it, it will ask you to crop it, although you can skip this step by clicking Skip cropping.
Now you simply need to place the photo on the work canvas. Just click the icon ✔ to place it where you want.
If what you want is to remove an element from your photo, choose the Eraser tool in the bottom bar and drag it over the area you want to recreate.
When you have deleted the object or person, select the Generation Frame tool and place it inside the area to be edited so DALL·E knows where to generate content. This frame cannot be resized; it is always 1024×1024 pixels.
All that’s left is to enter a text prompt in the top bar with which DALL·E will generate the edit. You should not write only what you want it to fill, but the concept of the whole image in general. Finally, click Generate.
After a few seconds, DALL·E will show you the result; in fact, it offers up to four different alternatives so you can choose the one you like best.
As you can see, you can hardly tell there was a cathedral in the image before. When you want to save the result, tap Accept.
Then, to save your new image, click the down-arrow icon that appears in the top right corner.
Combine two photos with artificial intelligence
Moreover, as mentioned before, DALL·E is also capable of generating content outside the image to expand its canvas. This can be used, among other things, to join two shots and thus create a kind of panoramic image.
To do this, tap the Upload image icon in the bottom bar and select your new photo. Once uploaded, place it where you see fit.
Center the Generation Frame between the two photos, write a descriptive prompt of the shots, and click Generate.
DALL·E will generate content between both images, joining them seamlessly and offering us 4 variants of the composite.
Since the Generation Frame is limited to 1024×1024 pixels, we’ll have to fill in the gap between the two captures little by little, but again, the effort is worth it given the incredible results it returns.
With this, you have at your fingertips a tremendously powerful and easy-to-use tool with which to retouch your photos without needing to learn Photoshop or any photo editing software.
.
