Florian Renaux crafts awe-inspiring realms that blur the lines between reality and fantasy. As a 3D artist, he possesses the remarkable ability to transport individuals to new dimensions, providing a sanctuary where minds can roam freely and imaginations can flourish uninhibitedly. Over the course of several years, Florian refined his distinct aesthetic until he forged his own unique universe, brimming with captivating imaginary landscapes and otherworldly wonders.
I started as a basic graphic designer, doing 2D, logo cutting, page placement, publicity, marketing and other things. Then one thing led to another, and I ended up here. What really got me into 3D was the artist Beeple, though nowadays what he does is not really my thing anymore.
I've always been attracted to nature. I live in a region where there are a lot of forests, this is also what inspires me. When you walk around, you're getting inspired by what surrounds you.
I always say that I transcribe the places I would like to see and experience myself. It's a bit like creating worlds that I dream of. They're places that exist, somewhere in my mind… I'm not a stargazer, what I do is a bit more fantasy, more dreamy.
I don't see all my creations as belonging to one shared world, I don't really have a guiding thread in my creations. It's more instinctive than that. Often my process starts on Pinterest, looking for an image in the universe that I want. Sometimes, I don't even know what direction I'm going in, I'll just type in landscapes or environments, fairly random things… Once I find an image I like, I start to build moodboards.
As soon as I start to have a more fixed idea about what elements will be in the scene, I deepen my research for related images. It could be a building, it could be a tree, it could be a statue. Sometimes I use Artificial Intelligence to generate more images, to inspire me further or bring out the ideas I have in my head.
When it's all defined in my mind, I go directly to Blender and start creating. My process has evolved: for about a year now, I've been the one to construct everything in my scenes. I had to teach myself to model because it wasn't necessarily my forte.
Once the vegetation assets are finalized, I move on to the rendering process, which is quite a laborious one. You have to render image by image. Before any color grading or post-production work begins, I upscale the footage to 4K and convert it from 24 to 30 frames per second, improving both the quality and the fluidity of the animation without needing to re-render everything in Blender. And you can render only what is volumetric: the clouds, the fog in your scene… So that you can have more flexibility in your post-production.
I work mainly on DaVinci Resolve, where the renders are compiled into a single video on which I can work, adjusting colors, brightness, and adding filters. You have access to a color palette to adjust the colors. Basically, you can accentuate or even add certain elements. For instance, I can try to remove or add some grain to mimic a real camera, to make the renders more natural. I also use a filter, a bit like Photoshop, that allows you to put a kind of a blur on the image… It gives the work a magical look. You really have control over a lot of elements when it comes to shaping the final image.
Another key aspect is optimizing the scenes. In 3D, you have to try not to have too many polygons in your scene, too many faces that compose your models, because otherwise your PC can crash. That's a really important element to learn: the process is the optimization of your scenes is essential, and it's something that you can't understand at the beginning.
As far as the work itself is concerned, I want to keep creating everything myself. To create a natural scene, you need texture, leaves, flowers, all these kinds of things. But I still integrate Artificial Intelligence into my work.
I have a model of a weeping willow and, thanks to new software, I just have to click one button to randomize it and obtain another tree that differs from the one I've already created. So you have some aspects that make creating easier. When you've already created a base, it makes the creation process easier, faster.
To create each asset, I start from the grass and then I build up to a tree. I'll start with a trunk, and then you instruct the program: I want branches, these branches will have other branches, which will hold the leaves. It's more of a node construction, you can randomize the whole thing with a button, and then it automatically changes, depending on the parameters you've set.
A lot of programs work with node systems. It's more practical because it's a view of the whole, of the construction of your model. For each tree, I use 3 models maximum, that are randomized in size and rotation. I also do this in the perspective of optimizing the scene, because if you have dozens and dozens of different tree models, it'll hinder your scene even more.
I think I would want the same reaction as mine: to step into these environments that I create, to dream and to say to yourself "Wow, I really want to be in this environment… to be in the place of the cameraman…" Quite simply, to give people this feeling of dreaming, a feeling of escaping reality.