Steven Cormann gave a brief talk on matte painting and the way he uses Megascans & SpeedTree and approaches his environments.
My name is Steven Cormann, I’m a lead matte painter currently working at Axis Animation in the UK. I also work as a CG generalist and concept artist and have worked for companies like ILM and Paramount pictures before.
I was born in the USA but moved to Belgium at a young age. As far as I can remember, I’ve always been drawing and been passionate about movies and science-fiction. Jurassic Park, Star Wars and Alien changed my life. Despite my passion, I never could find a way to see it as a possible career option until I was 30, which was roughly 5 years ago. After high school, I started a career as a drummer in the music field. I played drums most of my 20’s and I was doing some Photoshop on the side to design album covers. Then around 2013-2014, I hit rock bottom and my passion for art and movies really started to come back and I strongly felt the need to do something about it. I realized the times had changed and that what seemed impossible back in the day was now a reality for many people, so why not me?
I learned about matte painting and how those guys were manipulating photos to create those epic environments for films. Of course, there is a lot more to it but it showed me a clear path I felt I could follow and succeed in. So I started to work as hard as I’ve ever worked for anything. I kept working and posting my work everywhere and slowly the job offers started to come, and that’s how I got into the business. I also integrated myself into the community and attended events like THU, IFCC and Industry Workshops.
About Matte Painting
I do many different things but I’m mostly considered a matte painter. Matte painting consists of creating virtual environments using various 2D and 3D techniques. It’s quite a generalist job as it requires a lot of different skills. The main idea is to create an illusion of something that is not there that works for the shot. So matte painting is something done in post-production of a movie, TV-show or game cinematic for example. It can be a simple sky replacement or it can be an entire shot from scratch to finish.
Approach t0 the Environments
It’s all about what works best for the given job. I still rely on photos and painting but I tend to push the 3D more and more because I love the kind of rendering it can give me. There is only so much you can do with photos when it comes to creating a photorealistic environment. With 3D, I have the option to light and texture the way I want it, and then have total freedom with how I want the camera to move. I can also re-use assets later on. And I simply enjoy the process of creating 3D environments and then adding that little extra touch of 2D on top of it.
Use of Megascans & SpeedTree
I love Megascans, it’s now one of the best resources available out there I think. It gives me an incredibly photorealistic result that I can adapt to any lighting situation and camera move. It’s like using photos but with the possibility of changing the lighting and angle how you want it. And now with Mixer, you can combine those scanned surfaces and create your own materials. Concerning SpeedTree, they have an amazing library that looks extremely photorealistic, I use it a lot for the same reason I use 3D scans, it just gives me a ton of flexibility when it comes to lighting than if I had to use photo elements.
Concerning modeling, I never worry at all about the topology, for example. That’s one of the differences between a matte painter and an environment artist. Usually, the 3D assets that I create can be lighter as I can improve them a lot with 2D on top of it, and I use the same technique for my concept images. So my modeling techniques are basic, just polygonal modeling inside Maya and some sculpting in ZBrush. I then create some blend materials using various textures from Megascans or that I create myself in Photoshop.
First of all, I always have a good reference, that is essential to me. The lighting needs to be realistic so you can’t have the sun coming from one direction and then add a sky lit differently, I see that very often. Even if you do something not realistic, the lighting still needs to make sense. I look a lot at what’s around me, I take a lot of photos of the sky, landscapes, city etc. every day. All that helps build a visual library and understand how natural lighting works. I usually try to find a photo that I will use as a plate and I then try to recreate the lighting of the photo in my 3D scene the best I can. Once my render is done, I then make everything come together in Photoshop, but I usually try to push the 3D lighting to 80-90% of the final result. When it comes to the cinematic look, I often have a movie reference and I study a lot about cinematography, cameras, lenses etc. It’s all about having a good eye in the end, I think.
Advice for Learners
The matte painting tutorials from the Gnomon Workshop are a great resource, that’s where I learned most of what I know when it comes to the use and principles of 2D and 3D techniques. LearnSquared has a few good tutorials as well, like the one from my buddy Maxx Burman. Other than that, it’s really your eye that makes the whole difference, so having references and looking at nature is really the key.
Steven Cormann, Concept Artist at Axis Animation
Interview conducted by Kirill Tokarev
If you found this article interesting, below we are listing a couple of related Unity Store Assets that may be useful for you: