Using content-aware fill with smart objects

[dkperez]

I"m not new to Photoshop, but I'm new to smart objects, so if this is the wrong place to ask, please move this question... Oddly, I can't even find any discussion on this anywhere...

I'm opening an image as a smart object, and I want to extend the background using content aware fill.



At present the only way I've found is to

extend the canvas
duplicate the smart object

rasterize the duplicate layer

select the part to fill (this is the area of canvas I just added)

do the content-aware fill

leave the selection on and copy/paste the selection so it's on a new layer

delete the rasterized copy of the smart object



There's gotta be a better/smarter/faster way than this... How are other people doing this?

 

[RTContent]

That sounds like the most efficient way to me. Ps can only manipulate raster layers in this manner so that's why you'll need to rasterize every time. May I ask what's the reason of keeping it as a smart object in your case.?

 

[iDad]

If it is necessary to remain smart object just duplicate the layer shut one off

 

[ibclare]

Correct me if I'm wrong, but I think if you double click on the smart object you can the use content aware in this editing mode and save it . . .

 

[dkperez]

I believe the reason for wanting to keep it a smart object is so that if I want to make changes later, I can double-click and be in ACR. Normally - before being convinced that smart objects are fabulous and by FAR the best way to edit, I'd have taken my NORMAL pixel-bearing layer, and just extended the canvas and used content-aware fill.

iDad, can you elaborate on your statement about duplicating the layer and shutting one off? Duplicating the smart object just gets me another smart object, and I can't do any operations like fill on that.

ibclare, if I double-click on the smart object, it opens ACR. Which, I think, is the whole reason for opening the image as a smart object. If, on the other hand, I convert the image to a smart object, then double-clicking doesn't open ACR, it just goes into the smart object layer as another "image"...

I'm not sure this whole smart object thing isn't more work than it's worth... It sure seems like, with so many Photoshop features, that smart objects "sort-of" work. They do some things, but are a source of a bunch of new hurdles and limitations to work around.

 

[ibclare]

Well, you don't want or need to use smart objects all the time, or even often. But the ability to make a group of objects into an SO can be quite useful. I also like to use them, not just to open and edit parts of them, but to be able to apply filters non-destructively and go back and change them if desired. Also you are applying a filter or layer adjustment to one layer or a group. Also, SOs can be transformed, enlarged or diminished and the original information is still there. You lost pixels when you make a n object smaller. If it's a smart object, you can revert to the normal/initial size.

BTW, I think I'm wrong on using content aware even inside a smart object . . .

 

[iDad]

I only meant to duplicate the smart object layer work on the copy do what you need to do save it as a finished product go back to the smart object if you have to. Basically what I'm saying is save each step along the way.