 Lighting and Rendering - At the beginning I wanted to try the new Mental Ray pass system, but after a while I liked the raw render from Maya so I decided just to use beauty pass, occlusion pass, one pass for light fog and some masks. I ended with 16 passes, mostly masks because I wanted to do separate color correction while still doing an image ready for animation without hand drawn masks or retouching in Photoshop.
Lighting was real fun and also time consuming... I think I’ve spent most of the time on this and on compositing. I rendered a lot of different versions with different colors and light rigs going back and forth between versions until we found a good combination of mood and details.
I used a technique I saw in a DVD from Jeremy Vickery (www.jermilex.com) on the use of Ambient Occlusion to simulate bounce/fill lighting: all you have to do is to connect the output of the AO to the ambient color of ALL the shaders in the scene. You have to think about it like this: the light color is going to light the non occluded parts, while the dark one is going to light the occluded ones... so, avoid pure black and pure white!!! I put some orangish tones to warming the image up.
After establishing the key + fill I started adding bounce lights for the sunLight and a lot of rim light to separate the objects from the background. This is a screenshot of the scene and a list of all the lights in the scene. I also added some reflectors with a white surface shader to give that cartoony look to the reflections on the glasses.
Compositing - This was the first time I tried to use Nuke and I really learnt a lot!
I rendered some passes as 32bit floating point openEXR (beauty, zDepth) to have more freedom and the rest as 8bit iff.
Nuke gives you the possibility to create new channels from images, so I used the colorID passes to put new channels into the pipe in order to select the various objects with just one click through the whole script. This was especially useful while doing color correction. Height passes and the light fog helped me to improve the atmosphere while the vignetting allowed me to concentrate the attention on the car, which was supposed to be the most important spot of the image. Here you can see a screenshot of the whole script with some explanations!
Conclusion - I’m really happy about writing this tutorial and sharing what I’ve learnt with you! I think that this has been a little like a production workflow because we had to get feedback from each other by mail... sending progresses and trying to understand each other just by reading. Maybe if we had been able to work side by side we wouldn’t have done all those tests, but I think that this has been a nice research and I’m just looking forward to start a new project :) If you have any questions feel free to write me! Thank you to everyone!
Luca Fiorentini
Blog: lucafiorentini.wordpress.com
Giovanni Lo Re
Portfolio: www.pixila.com/portfolios/portfolio/?artist=1424
|
