Like everyone, I’ve been having a play with Adobe’s new photo editing tool, generative fill. Using AI it is now a short leap from imagination to realisation. Ruth and I visited Richmond Castle yesterday and as I looked at the tower my brain screamed “I Have The Power”! And even in a campsite with mobile phone data, it seems I do. After the edit, I added heavy colour grade in Lightroom and away we go.
Read more: Richmond Castle Grayskull | Trying Generative Fill
My first impression of generative fill is that it is amazing at removing objects. People? Gone. Information stand? Gone. Bench? Gone.
I was messing about for around an hour and only with this one image. I tried a few different things that were less successful. “He-Man” as a prompt produced all sorts of modern macho men. “Anglo Saxon warrior” gave all sorts of different fantasy genre inspired Barbarians. A rack of battle axes and swords goes against Adobe’s rules so it refuses to do it. Even robots have morals.
One of the features causing the biggest stirs online is the ability to make an image bigger where AI generates the rest. When I tried it with this image, the colour of the clouds didn’t match. I expect that is because of the nature of the image.
Here are some other non-photoshopped images from our trip. All in all, a lot of fun was had.