Participant

Answered

Replace image, keep all edits from original

Forum|Forum|2 years ago
October 19, 2023
3 replies
4882 views

Hello there,

So I'm working on an image in Photoshop. After finishing editing (mostly selecting and deleting backgrounds, fine-tuning masks, etc...), I realize I wanna replace that image with another version of that same image, but keep all fine-tuned edits and post-production from the original, on the new file. A simple replace content does not seem to work it just places the new image, without copying the edits. Neither does copy layer style, I guess because the edits are not separate adjustement layers. Any other way to work around this ?

many thanks,

MP

macOS

This topic has been closed for replies.

Correct answer Conrad_C

It takes some planning before starting the edit. You want to do as many adjustments as possible using parametric/nondestructive methods stored in separate layers, so that all you have to do is drag those layers to another image and they adapt to the new content. This works OK when all of the edits can be done as adjustment layers, layer effects, or Smart Filters, all easy to duplicate to other images. But any edits that permanently alter pixels on the original image cannot be copied to another image…they’re “baked in.”

Some of the techniques that simplify bulk edits are easier for advanced users. For example, Blake Rudis pointed out that although a manually created (rendered to pixels) luminance mask cannot adapt to another image, an adjustment limited by BlendIf luminance controls will adapt to another image because it’s based on levels, not pixels.

If your images start out as raw files, then a better strategy would be to start by doing as many edits as possible in Adobe Camera Raw or Lightroom/Lightroom Classic. The reason is that those applications are fully parametric and nondestructive, so practically all adjustments are easy to sync or paste to other images, and many of them auto-adapt.

For example, if you retouched skin in Camera Raw or Lightroom by generating a People AI mask limited to skin, when that mask is synced or pasted to 10 other images with slightly different poses, the application AI re-analyzes the poses in the other images and auto-adapts the people skin mask for each of the other poses, saving much time. Photoshop cannot do this yet. Similarly, Camera Raw and Lightroom have proper built-in luminance range, color range, and now Point Color masks and those easily auto-adapt to other images too.

Camera Raw and Lightroom can’t do everything Photoshop can, so after the first round of quick bulk edits in CR/LR, any images needing further work could be moved to Photoshop for finishing.

Conrad_C

Correct answer

Community Expert

It takes some planning before starting the edit. You want to do as many adjustments as possible using parametric/nondestructive methods stored in separate layers, so that all you have to do is drag those layers to another image and they adapt to the new content. This works OK when all of the edits can be done as adjustment layers, layer effects, or Smart Filters, all easy to duplicate to other images. But any edits that permanently alter pixels on the original image cannot be copied to another image…they’re “baked in.”

Some of the techniques that simplify bulk edits are easier for advanced users. For example, Blake Rudis pointed out that although a manually created (rendered to pixels) luminance mask cannot adapt to another image, an adjustment limited by BlendIf luminance controls will adapt to another image because it’s based on levels, not pixels.

If your images start out as raw files, then a better strategy would be to start by doing as many edits as possible in Adobe Camera Raw or Lightroom/Lightroom Classic. The reason is that those applications are fully parametric and nondestructive, so practically all adjustments are easy to sync or paste to other images, and many of them auto-adapt.

For example, if you retouched skin in Camera Raw or Lightroom by generating a People AI mask limited to skin, when that mask is synced or pasted to 10 other images with slightly different poses, the application AI re-analyzes the poses in the other images and auto-adapts the people skin mask for each of the other poses, saving much time. Photoshop cannot do this yet. Similarly, Camera Raw and Lightroom have proper built-in luminance range, color range, and now Point Color masks and those easily auto-adapt to other images too.

Camera Raw and Lightroom can’t do everything Photoshop can, so after the first round of quick bulk edits in CR/LR, any images needing further work could be moved to Photoshop for finishing.

M

MP1650Author

Participant

Many thanks for your detailed reply. Yes I guess I'm starting to learn the planning thing. You correctly pointed out Lightroom. I am actually working on one High resolution photograph, that was edited in Lightroom, then opened in Photoshop, and further edited. Once the post-production was done in PS, I realized I needed to do some further editing on the same photograph in Lightroom (that allow for certain features that PS doesn't have), then wanted to re-import the photograph (so strictly the same photograph), into PS, but keeping the post-production already done in PS previously. That didn't seem to work, also I realise that although PS and LR communicate with each other, it is limited, and you cannot go back and forth continuously, as, as was pointed out here, some edits in PS are destructive, whereas in LR they are not.

D Fosse

Community Expert

Yes, round-tripping back and forth between PS and Lr can be a dangerous trap. I always advise people to not do that until they have a full understanding of the differences between these two applications, and crucially, the different ways they treat the data.

For instance, Lightroom does not support or read Photoshop layers. Any layers in Photoshop will be ignored and lost in Lightroom. What Lightroom works on is a flattened composite inserted in the file by Photoshop - as long as you have "maximize compatibility" set in PS, which you have to if you want the files to be visible in Lightroom. In short, all you will ever get out of Lightroom is a flat file.

On the other hand, all Lightroom adjustments are text instructions saved separate from the image data, so always fully re-editable and non-destructive.

If you want/need to use the Lightroom adjustments on an RGB file from Photoshop, a much better way is to use the ACR filter in Photoshop. It accomplishes the same thing, but does not destroy the layer structure, and can also be applied as a smart object and/or with masks, so much more flexible.

But generally, and as a working practice, Lr > PS should be considered a one-way operation.

And as Conrad points out, it always pays to do as much as possible at the raw stage in Lr/ACR. Not only because it's non-destructive, but because a raw file contains a lot of data that necessarily has to be discarded when encoding into an RGB file. The linear raw data can hold a dynamic range of up to 14 full camera stops without clipping, while a gamma encoded RGB file is normally limited to 7 or 8 stops. A big part of raw processing is to map those 14 stops down to 8, using highlight recovery and so on. You can't go the other way.

Trevor.Dennis

Community Expert

Wow. That's a really big ask. I don't suppose the image is a background layer with a stack of adjustment layers above it?

Or a Smart Object with a list of filters like Camera RAW underneath it? I don't suppose you have Preferences set to record a history log of the steps you took with the original?

M

MP1650Author

Participant

Thanks for your reply, no I haven't recorded a history log, but I will look into that the next time. Many thanks.

D Fosse

Community Expert

If you're not using adjustment layers, but direct pixel adjustments, then no. That's like retrieving the eggs from a cake.

Masks can be dragged between layers or copied between documents.

Trevor.Dennis

Community Expert

That's like retrieving the eggs from a cake.

By @D Fosse

A nice annalogy Dag. I don't want to be unkind to the OP, but it reminds me of an old workmate — and bear in mind that we worked in the Test Instrumentation department of Ford Product Development, so he was not without technical know how. Anyway, this workmate thought that the way that TV stations could tell how many people were tuned in at any given time, was by how much transmission power was being absorbed. The more people that were watching, the more they would need to turn up the signal. That's a true story!

D Fosse

Community Expert

@Trevor.Dennis I like that theory, makes perfect sense, doesn't it? 😉 Crank it up to eleven, guys, we got a big crowd tonight -

Sign up

To post, reply, or follow discussions, please sign in with your Adobe ID.

Sign in to Adobe Community

To post, reply, or follow discussions, please sign in with your Adobe ID.

Scanning file for viruses.

This file cannot be downloaded