Is it feasible to mix 3d images with fine art photography?
I'm working on a couple of fantasy photo projects and find there are some limitations with what I can do with the camera and Photoshop. It seems that some of the "very realistic" models might work well in some of my projects.
I would appreciate any feedback about the concept of incorporating subjects generated within Daz into my lanscape photos along with human models. I'm sure that I'd have to desaturate the Daz model some and blend the skin to have it look like it belongs in the image.
Can I generate a .PDF format output from Daz?
Thanks in advance.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
Do you mean .psd? No, but Tiff or PNG will preserve an alpha channel.
Opps... yes, I meant .PSD format.
Thanks for the info about TIFF
Yes. I meant .psd. I guess the TIFF or PNG work work fine.
thanks for the info.
Oops.... Yes I meand .psd. Thanks for the info about the TIFF and PNG format. Those should both be useful.
It would be a fun concept to play with with some of the more photorealistic renders. Facegen artist is one way to go for photorealism and of course you would need a graphics editor such as Photoshop or Gimp. Personally, I prefer a slightly painted look so if I use photos, I have to simplify them first. Good luck!
I have used my own photographs in images.
https://www.daz3d.com/forums/discussion/comment/2637826/#Comment_2637826
This from farther down the page.
https://www.daz3d.com/forums/discussion/comment/2638311/#Comment_2638311
Certainly! Stock is stock, no matter if it comes from a camera or a render. You can blend them in all kinds of ways and make them look good together. I've seen quite a few users do this and I enjoy mixing the two myself. It really boils down to good planning (ie matching the lighting, good render settings, choice of images to blend, etc) and photo manipulation skills.
Here are a few I did for fun over the past year:
I should have said, mine are straight renders with no Photoshop.
I suggest reading up on on the uncanny valley
https://en.wikipedia.org/wiki/Uncanny_valley
Depending on how realisitic you require your models to be, it might require substantial editing in photoshop. Some of the genesis 8 models are pretty realistic, but I find that their eyes, feets/hands and facial bone structure often gives them away as fake. Human features are filled with inperfection and nuance while many 3d models, although having photographic skin textures, tend to be "too perfect" in symetry and definition. There is some dforce clothing I would consider "photo-realistic". CG hair is also a dead giveaway. I have yet to see a CG hair style which can pass as real hair. Even hair rendered in Disney farms falls short. Rendering your models at the right perspoective angle of your composition is also important- off by a degree, human eyes can pick it up. Creating a realistic light that matches the rest of your composition as well as shadows that interact properly is also of utmost importance..
Yes, it is feasible, depending on the effect you are going for. I do it all the time, actually. There are three main tricks to making this work:
Here's an example that worked out pretty well. I did this one in Carrara instead of Daz Studio, but the principles are the same. One useful trick is to use the image background for image based lighting (IBL). In this example, I took the background image, enhanced the contrast, converted it to an HDR image, and used it as an HDRI based environment light. As a result, the background image itself is acting as a light source and casting light into the scene. This helps with color and light matching to give the illusion that the 3D character is actually part of the scene. (It's not the only light source in this scene, of course. There are some spot lights and point lights being used for special effects.)
Btw, I'd also suggest that not only is it feasible, it's actually desirable to mix 2D and 3D assets whenever you can get away with it. After all, if the background in the image I posted had been an actual 3D object with ocean spray, tides, 3D clouds, 3D lightning effects, etc? That image probably would have taken 4 to 10 times longer to render. By using 2D assets when possible, you can greatly reduce your render times.
Great work, Llynara! #3 and #4 are particularly great examples of how mixing 2D and 3D assets can produce a very convincing illusion. And no doubt, it saved you a lot of render time too.
Thank you! The first two were actually done for Sketchfest, and I only had an hour to work on them from start to finish. I spent 15 minutes setting up the render, 30 minutes rendering and 15 minutes doing postwork. The others were not timed, so they look better.
One trick I've found very helpful for blending renders with photos is to use layer styles in Photoshop to fake rim lighting. It lets you soften the transition from render to background photos. I learned it from this blog post and use it often, tweaking color, intensity, etc as needed and then saving as presets: http://www.tipsquirrel.com/adding-rim-light-photoshop/
Thanks clmoonriver. I haven't even thought of that but making the render more efficient would be a nice bonus of combining the 2D stock images.
Great work Fish. I love those images. thanks much.
I'm taking an online course by a Photographer (Brooke Shaden) which focuses on Fine Art Compositing and, as you suggested, she spends a lot of time teaching how to corectly blend the images and to get the lighting right. If it's not done right, it doesn't look real at all. thanks much.
Thanks so much for the great info Llynara. I'm particularly inspired by the image "Ready" as the color and texture that you selected for the 2D and 3D images meld so well that there is a great sense of realise (besides just being beautiful).
I like to play with 3d in 2d, too! I found a very good tuto by Dreamlight https://www.daz3d.com/cg-live-footage-mastery
For me it works best to set the image as background and search for an HDR that matches this pic. If I have the correct perspective point and camera adjustment (the hardest part for me)
I switch the Iray preview on and turn the dome as long as the shadows of my 3d content are correct. Than I render without the background and save it as .png
In PS I put it together than and so is it possible to do some alterings in colour etc... If you like, here are three examples:
http://fav.me/d9gjh8y
http://fav.me/db3axsw
http://fav.me/dc3zswy
In every case you need one or more shadowcatcher. How to use them here a brilliant tuto bei MEC4D https://www.youtube.com/watch?v=6vaUXdZzy0c
It's a great way to place your characters into environments you don't have in your 3D content. For instance on the attached, the forest path backdrop is from a royalty free computer wallpaper site. So I just had to get lighting and shadows aligned for my characters and they're walking in a landscape I couldn't re-create with my 3D content assets.
Oh, and I do my 2D work using GIMP, a powerful free tool available from http://www.gimp.org/
Almost every single image I do is a composite of 3d and 2d. I don't think I could do the art that I do if I didn't do it that way. Layers are super important as it can take more than one layer to blend everything together seemlessly. As mentioned above, thed direction of the lighting is super important.
An example of the raw render and the finished verison.