94Kviews
21 Images That Reveal Hidden Scenes Of Famous Images, Made By Twitter User Who Tested The Newest Photoshop Feature Interview
94Kviews
There is no doubt that AI is developing rapidly and every day we see how advanced it is already, just to be even more shocked the next day to discover some more possibilities it has to offer. The points of view people have about this technology are rather diverse. Some of them are amazed and excited to see how much it helps in their everyday tasks, while others hate it for sourcing its “skills” and inspirations from other, real artists. There are also some people who maybe have heard something about it, but they do not care to explore this topic.
More and more software has started introducing features based on AI. Adobe is no different. Very recently, Photoshop released the new tool, AI Generative Fill, allowing, for example, removing unwanted objects from the background, adding objects to the image, or extending the canvas. Twitter user StepnEurope decided to test the newest feature and use it for extending his favorite memes and photographs. We must say, the effects of this experiment are quite impressive! Let us know in the comments below what you think and if you have yet to play around with Photoshop’s AI Generative Fill.
More info: twitter.com | Instagram | Facebook
This post may include affiliate links.
Disaster Girl
You can tell it's AI because of the blur also the situation has escalated from one house to the whole town.
Bored Panda reached out to Beka Zakaidze, the person behind the StepnEurope Twitter username, to find out more about the images he created using the newest Photoshop feature. First, we wanted to know what inspired him to apply the Generative AI Fill technique to popular images and memes. Beka told us: “I've been using AI tools from day one and have always been a big fan of generative AI tools like Midjourney and MOOAR. As soon as I saw the capabilities of the Photoshop beta, I wanted to test it on images that we are all familiar with. I couldn't think of a better choice than memes, so I grabbed my favorites and began experimenting.”
Think About It
Lunch Atop A Skyscraper, 1932
Next, we asked Beka if he stumbled upon any challenges or limitations he encountered while using the AI feature on these images. He answered: “The funniest part is, despite using Adobe tools for a decade, the Generative AI Fill tool is incredibly easy to use. You simply highlight the empty area and click 'generate'. I didn't retouch the final photos to showcase the true capabilities of Adobe's Firefly.”
On the contrary, we asked if he encountered any challenges while modifying the images. Zakaidze answered: ”Not at all. This was genuinely a one-click fun experience. I'm not even sure why it's in beta; I can hardly imagine what the final version could do better.”
Hide The Pain Harold
Harold's body was cut in half one day when he volunteered for audience participation at a magic show. Now he rests comfortably on his favorite chair and a huge pillow.
We were wondering if there were there any specific memes or images that stood out to Beka as particularly intriguing or enjoyable to work with. We found out that: “The first one in my thread, 'Disaster Girl,' was the most fun to work with. So many details were unveiled; I felt as though I was always looking at the extended version.”
Side-Eye Chloe
Afghan Girl (Sharbat Gula) By Steve Mccurry
Lastly, we asked Beka to share some future plans or ideas he has in mind to further explore and innovate with Generative AI Fill. He revealed: “I've tried this tool on iconic photos and Einstein's photo came out particularly great. As I'm using these tools daily, I plan to periodically share my work.”
Bigfoot, Patterson-Gimlin Film
Ancient Aliens
Neil Armstrong, Apollo 11
Exit 12
American Troops Approaching Omaha Beach In World War II
These suck! No imagination to make them fun..just a.i. being boring
These were all interesting but very inaccurate
That's kind of the point. AI can make something that looks vaguely right, but that doesn't actually mean the AI has any clue what it's doing. They're not actually in any danger of replacing us any time soon.
I'd be interested to see how actual humans would complete this task and comparing the results.
What was the point? It didn't add any value to the viewer
These were all interesting but very inaccurate
That's kind of the point. AI can make something that looks vaguely right, but that doesn't actually mean the AI has any clue what it's doing. They're not actually in any danger of replacing us any time soon.
I'd be interested to see how actual humans would complete this task and comparing the results.
What was the point? It didn't add any value to the viewer