Newbie trying to achieve the Google Pixel "look", please critique

[Sciencenerd]

I also like to thank everyone, I am learning a lot, although some of the info is over my head.

I maybe should have clarified this in my OP, I am NOT trying to replicate exactly what Google is doing in GCAM, I realize that there are bursts of shots and lots of computational photography involved. Nor do I want to do bracketing for HDR photography with my OM-D (except in very select cases). I just want to use the natural dynamic range of the OM-D camera to get a Pixel-like look from a single exposure. My limited tinkering with this in DT seems to show that the OM-D actually has plenty dynamic range, that bests the shadow detail from multi-shot exposures of the Pixel even in a single exposure.

What I am rather looking for is a simple workflow that expert post-processors, like you guys, think makes sense, so I can give the OM-D photos a little more of that Pixel “look” that seems to be pretty pleasing to many people. Hope that makes sense.

What I am learning is that a lot of other posters seem to use tone equalizer to achieve the DR compression effect, rather than filmic, before adding back local contrast.

[Entropy512 (Andy Dodd)]

@Sciencenerd

I just want to use the natural dynamic range of the OM-D camera to get a Pixel-like look from a single exposure

As I mentioned, if you do want to do that:
Export two (or more) different exposures with different exposure shift in your raw processor
Feed the result to enfuse

That applies the elements of Google’s pipeline that are relevant to a single exposure from a camera.

(I’ll try to do an example later this week, the timing is bad at the moment.)

[age]

P5150281.ORF.xmp (10.1 KB)

[mikae1]

Sorry, but not an attempt to replicate the Pixel look (which I find rather lacklustre).

Newbie trying to achieve the Google Pixel look, please critique.ORF.xmp|attachment (61.9 KB)

[Sunhillow (Hansgeorg)]

My try using RawTherapee
Tone Mapping, Haze Removal, HSV Equalizer, Wavelets → Clarity
Overprocessed to my eyes

P5150281_RT-1.jpg.out.pp3 (15.6 KB)

[Thomas_Do (Thomas)]

My Pixel version:

P5150281.ORF.xmp (8.2 KB)

How I would do it:

P5150281_01.ORF.xmp (8.2 KB)

[saknopper (Sander Knopper)]

Must say I quite like this image, here’s my attempt:

P5150281.ORF.xmp (8.6 KB)

[priort (Todd Prior)]

Its funny I had a Lumia phone which in its time took great photos for a phone…There was a certain look to the contrast and the tone mapping…some of which reminds me of the google look…messing around when the old tonemapping module was available (now depreciated) I was able to get some settings with that module that would impart something similar…I thought I had an old preset so that I could access it but I can’t seem to find it so I would have to go back and install an old version or whatever and its not really worth it but it was interesting …for fun I may dig it up and see if my memory is correct or I am dreaming

[qmpel (blends colors 🎨)]

Sorry for the Bump, but I would like to add my 2cents to this topic.

I found this a really nice challenge - something to train your eyes, spotting the color, sharpness and contrast differences and, of course, using the tools to minimize them.

I tried really hard, to get a very good match between my edit end the Pixel 3a reference, but I failed because of 3 reasons

• the reference photo has same subtle color gradients in the greens or the blues (in the sky). These are not decisive for the look, i think, but make the exact matching very hard
• the OMD picture is a bit blurry in the foreground (the bushes for example)
• there seems to be a different lighting in the near foreground (the tree, its shadow and the plastic cover)

So after trying several hours with masks and gradients, I ditched this approach and tried a more imperfect but quicker edit.

First of, here is the end result and I think it matches pretty good in the overall contrast and colors, the sharpness and local contrast.

As to what I’ve done… I found the most important bit is the dynamic range compression, while maintaining or enhancing local contrast, which can be done (like @s7habo and others have shown) with two instances of tone eq - this gives really good control over where exactly to place your contrasts.

Second, I looked for the colors. I made the sky more bluish and kept clouds mostly neutral with a little bluish cast. Not sure, how the Pixel 3a does this, but I tried to match the colors with the channel mixer, as I like to use this module. I kept an eye on the greens, and balanced them between cold and warm to match the reference.
To match the blues I used the input blue in all three color channels of the mixer and for the greens, I mostly used the input red/green in the blue color channel.

At last I tried to match the sharpness with multiple instances of diffuse or sharpen. This is pretty standard for my editing process and here i used a lens deblur, dehaze and sharpen demosaic, with my personal adjustments and balanced out to kinda match the reference.

So, to wrap this up… I thought I’d add a little information about my thought process, which may or may be not useful for the op (or other interested people), because I think it’s a common theme for, err, newbies, trying to match the style of their previous photos with the phone.

Thanks for the fun and the learning experience on my side, @Sciencenerd!

[dqpcoxeas]

My version…

P5150281.ORF.xmp (19.6 KB)