In my previous blog post Panic at the packaging factory? I mentioned that I had used Photoshop to fill in the gaps in the upper corners of an image background. Due to the short time I had to finish the job, I used the eyedropper tool to sample the background colour and then painted in the corners using a brush. This worked fine because the background was almost white, and there was no significant colour gradient next to the area I needed to fill.
I was subsequently asked to create some similar shots and, with more time to do the post-processing, decided to try the Generative Fill option in Photoshop instead. This is a relatively new and powerful update, but what I wanted it to do was just the tip of the iceberg of what it can potentially add (or take away).
Generative Fill is available from Adobe Photoshop version 25.0 onwards.
With the image open in Photoshop, I needed to select the area with the missing background, so I chose the lasso tool and traced around as close as possible. As I completed the selection, the floating Contextual taskbar automatically changed to the Generative Fill taskbar.
Note: If you can’t see the Contextual taskbar then click on Window in the Menu bar and look towards the bottom of the long dropdown list and make sure that Contextual Task Bar is ticked.
I pressed the Generative Fill button and then the Generate button. By leaving the “What would you like to generate” box empty, Photoshop automatically filled the area with content based on its immediate surroundings.
It takes a little time but this is indicated by a progress bar.
It will then apply the generated content onto the image, but also give you another two variations in the Generative Layer panel to the right. You can press Generate again if you are not happy with the initial three offerings.
The generated image is applied as a layer so you can always go back and change it later if you want.
Generative Fill uses Artificial Intelligence (AI) to decide what to add to the image. In the case above it was used as a shortcut to fill in an area of missing background, but, of course, it could have been used to add something else entirely, maybe something that may be seen as being slightly dishonest. This is the conundrum facing photographers: what is acceptable and what is not? Of course, this is not a new phenomenon; the Cottingley Fairies photographs from 1917 are early examples of ‘generative fill’, albeit without sophisticated software.
Discover more from Ironbridge Images
Subscribe to get the latest posts sent to your email.
Ironbridge Images 