warning: all target frames are blank. The Content-Aware Fill layer generation has been aborted

[brian taylor]

After Effects ver. 17.7.0 (build 45)

Mac OS Big Sur 11.2.3

I've created a tracking mask for a quicktime animation. Mask is set to SUBTRACT.

Layer is selected.

When I try to generate Content Aware Fill Layer, it gives me a progress bar, then the above error message.

See the attached screenshots.

Have tried in composition and layer view, same result.

Have trashed the preferences folder, same result.

[Mylenium]

This will never work. You are using the wrong approach. Of course this cannot be filled with anything AI-based, because there is simply no pattern to detect. And why would you even? The whole area is just plain black and easily can be covered up with a small black solid. Again, you are going about this the wrong way.

Mylenium

[Cris-media]

Same case, same error message, and the video area is filled with text that needs to be removed. Can you offer a solution?

[the52films]

 Same Error. See I am Completely new in the affter effects. Completely means Completely. These is my first time ever installing & creating a project in the Affter Effetcs. Please tell me the solution in which i can understant. Please 

[Rick Gerard]

For Content-Aware fill to work properly, the object you want to remove must move away from the background below the first frame, revealing the new background. For example, a shot of a person running down a road from the side where the camera follows the runner is a good candidate for CAF to work because the background behind the runner in frame one will not be hidden in frame ten because the runner has moved. There will be details that were once hidden that can be used to fill the hole in frame one created by the mask.

If the runner is running directly toward the camera, there will never be any details behind the runner. Even the new AI-powered Generative Fill in the latest version of Photoshop would have an impossible time creating a consistent background replacement in an image sequence. I've tried. Removing a runner from a 3-second burst from my Sony A1 that will shoot 30 frames per second at the maximum burst rate was impossible. I would bet a pretty good image from the first shot, but the AI-generated background on the second would always be different.

The only option you have is to motion track or motion stabilize the object in the footage you want to replace, create, or clone, or in some way fix the background, then animate that fill as the footage changes so that the shot is fixed. I had that kind of problem in some shots I worked on for a feature film. I had to remove a tattoo from the neck of an actor as he moved around the set, in and out of shadows. It took Mocha Pro's mesh tracking + Lockdown + some puppet pin animation + several weeks to fix about a minute and a half of screen time. A few years ago, the project would have taken a few months because each frame would have had to be painted by hand.

[the52films]

Thank you for your response! I'm still having trouble understanding. Is there an alternative explanation or perhaps a visual demonstration, like a video or assistance through AnyDesk? Feel free to reach out to me on Instagram @ThE_52_films