Let’s find out how these new generative AI tools will undoubtedly be coming to video editing soon.
Like many people who work in the film and video industry, I have many friends who work in design and creative services. Believe me, while AI is wreaking havoc on the world of professional video, it’s also completely destroying the creative design profession. (After all, as we have proved, indeed some Elements of video editing utilized by AI probably will never really replace. )
Still, after hearing the news that a new generative fill tool is coming to Photoshop, let’s see what new generative AI tools Adobe (and other NLE makers) will soon be releasing in apps like Premiere Pro and After Effects. I can’t help but wonder, not just whether. That’s the key takeaway from this news. It’s no longer a matter of “if”, it’s simply “when”.
Let’s dig deeper into this new Generative Fill tool for Photoshop and explore its potential to be a great leader among other generative AI tools expected to shape the future of video editing.
Generative Fill in Photoshop
Released as a new beta tool in Photoshop, Generative Fill is designed to help creators “bring their creativity to life” by allowing them to beautifully update their images with a simple text prompt. Just like we’ve seen in other generative AI apps like MidJourney and Runway ML (This is already a great case study for generative AI video editing.), this new tool is just one of the first in a suite of new AI-powered tools and features from Adobe.
In particular, Generative Fill aims to add, extend, or even remove image content in a non-destructive way, simply by using text prompts. When done right on the right projects, the results can certainly be very impressive, as Adobe’s AI attempts to match photorealism with additional features that can be rendered in seconds.
However, Adobe is taking small steps in this deployment.After presentation Adobe Firefly at NAB, there’s been a lot of speculation about what AI-powered tools the company plans to incorporate into its apps first. Generative Fill is certainly very impressive, but really only intended to supplement or add to your project. It is not intended to completely replace design or photography work.
Upcoming generative AI tools
Also, as we begin to dig a little deeper into this new generative fill tool, it’s interesting to note what this tool can do beyond headline-catching elements. This new tool is actually a little more powerful than advertised in its marketing. In addition to the ability to generate backgrounds for objects with text prompts, you can also generate objects (and possibly even people) while augmenting images and removing objects (to be fair, Adobe has been offering for some time).
Here’s everything Adobe promises you can do with generative fill today.
- object creation: Select an area in the image and describe what you want to add/replace through the text prompt.
- generate background: select the background behind the subject and generate a new scene from the text prompt.
- extend image: Expand the image canvas and select an empty area. Generating without prompts creates a harmonious extension of the scene. Generating with prompts adds content to the image while extending the rest of the scene.
- delete the object: Select an object to remove, generate without prompting, and Generative AI technology will hide the object.
- Plus… generative fill is incredibly versatile. Experiment with outlandish ideas, come up with different concepts, and instantly create dozens of variations.
The real question here is what Generative Fill AI will enable in tomorrow’s video editing.
video generative fill
I believe that if Adobe can roll out this Generative Fill tool for Photoshop today, they could roll out a similar tool for Premiere Pro tomorrow. The limit is not necessarily on the AI side, but of course on the processing side. Rolling out Generative Fill as a beta with Photoshop is considered a good way for Adobe to gauge interest and computing power before introducing generative AI tools into other programs.
But as we said, Premiere Pro and After Effects have had AI tools and other smart features for years. Sure there are some tools that are cool and useful in certain situations, but I don’t think AE’s object removal tool or content-aware fill tool destroyed the internet or the industry. Rather, it’s fun to experiment with and use occasionally, but it’s not widely used in most editing pipelines and workflows.
Even if Generative Fill is released for video editing tomorrow, there aren’t many examples of it being useful at macroscale. And even so, the amount of microcontrol you have to do per frame is probably possible, even if he converts, say, a portrait video taken with an iPhone into a 16×9 composite, but it’s a considerable amount of time. It takes Consume regardless.
Still, it’s probably comforting to know that Adobe (and everyone in the field) is doing a lot of research to find out how best to deploy more generative AI into their video editing products and services. increase. And if you think this fad might pass, unfortunately you’re wrong.
What do you think of this new generative fill tool for Photoshop? Do you think you’ll be using something similar when you get to Premiere Pro? Let us know what you think in the comments below.
