JamieStefankoPhotography

And that would be perfection, thank you! How did you manage this?

Seriously. I had to rush and get a family photo done for a school project for my daughter, so I did this with a timer. Only one face swap, I was so proud! And then I saw my stupid shoe. I only have one image with the shoe on my foot, but I can't get it to look right (it was taken when I was sitting down). Any suggestions? 

Damien - I somehow missed the notification for this, just seeing it now. Fantastic job, as always. Will take a closer look after dinner, and give it a try on a few other pictures - thank you.

And you suggest just careful masking around the subject?

Yep, that's what I'm looking for. I really don't mind a dark grey instead of blue either (meaning, grabbing from the color in the background) if it makes it work easier. This is why I ended up resorting to that stupid Quick Select.

Sure. However, there's one pixel layer I'm having trouble getting rid of, and that's the separate one of my son away from the background (and even this isn't perfect, but better than what I'm coming up with). The most careful of masking isn't really helping me with this. 

I did. I was literally just trying different things to see what would work, it's not my normal workflow. 

And I'm wondering if the issue is that I used solid color layers instead of gradients. 

Sure - I used the dreaded Quick Selection tool, but believe me, it was a last resort. There's levels work above not shown, but that's mostly on his face.

 

Just trying for a realistic looking background replacement - the joys of not having a studio and backdrops. Not looking to print it, but the banding was pretty awful - think the noise layer is at 40% here. Anything I can do to improve this? SOOR and edit.

I can say, it does look a bit magenta. So, maybe add some green?

No, more of a matte-ifying effect, likely due to the cotton rag paper.

I should test it against that metal print profile from h+h, but I assume I should get test prints back first, right?

With soft proofing off, my current setup using Method 2 looks good, so yay, haha. Wish I had some h+h samples to hand, have to go remedy that, but happy with Musea at least.

As per Method 2:

 

Aaaand this is 6500K. Ooof.

 

Sooooo, skip this and go to method 2? Like, I don't even know, ugh.

3 minutes ago, Samantha LaRue said:

You'd go with the top one.

 

But holy cow! Why is your monitor so bright? There is no way that matches your prints, right?

I just reset it to factory defaults, lol. So, it looks completely wonky right now!

This looks a bit different from the instruction screenshots - I have three values per box. 

So of course I did the first step (not even calibrated yet, just set it to 5000K) and it almost matches perfectly, lol. Of course, right?

Alright then. It's a very overcast/rainy day, still better off waiting till nighttime?

10 minutes ago, Samantha LaRue said:

This might be why your metals seemed overly magenta too! (they looked fine to me)

Sounds like you'll need to calibrate again. Do you have the instructions handy for your device?

That's likely exactly it. I had saved the profile, but maybe ultimately upgrading to Windows 10 did me in somewhere, sigh.

I think I should be following this.

http://www.damiensymonds.net/cal_i1DP_pc1.html

Bah, they're a bit off from Miller's too - I upgraded my PC less than a month ago, so it's a new calibration.

Before I ordered prints, I soft proofed them with their profile, its that where I went wrong? I don't remember how I did it with Miller's, it was ages ago. 

I've been using Millers for some time, and I was calibrated for them. I'm switching to Musea (though I'll also be using one other lab for canvas prints). I haven't touched my calibration other than my standard monthly recalibrate, and I received my test prints from Musea yesterday. I have their ICC profile on softproof, and when comparing, theirs is definitely warmer. So, now I need to match my prints, right? I confess, when I first calibrated to Millers, I didn't have to change a single thing, haha. Do I try to do this through the X-rite software, or what?

Also, if you use two labs, do you just choose the lab you use the most when calibrating?