Render Test

We're now 6 weeks into the school year and I've started off my rendering class at a more reasonable pace than in past years. The first series of lessons was to get into the use of mental ray for rendering, however we're not yet using advanced lighting such as Final Gather or Global Illumination (yet). Along with that, the first month was a big introduction to linear workflow. Simply using spot, area, point, and directional lights, we've been looking at how to direct the eye, cast realistic shadows, and look at light colour temperature all while working on our basic material knowledge.

In this still life example I've set up, the students were allowed to use render layers in order to break up background from foreground, and use ambient occlusion to augment the contact shadows. This scene of mine is 4 layers of depth, along with their respective AO layers. In the image above on the left you can see the original layers and what it looked like straight out of Maya. On the right is the Fusion tweaking I've done with masking and colour corrections to try for something a bit more photographic. I find the glass and chrome table on which the breakfast tray is sitting just a bit confusing to look at, however as with so many other projects, it's time to move on. As usual, feedback is welcome.

An Update

Here's a small update to the teacup image. I'm getting it to the point where I can use it for a lesson. It's coming along but I still have a small list of things to fix before I can do this with my class. I'm getting happy with it though.

Just A Test

On vacation in Europe this summer, I took this little picture. Nothing spectacular about it, but as a lighting test to get my class into advanced mental ray shaders and lighting, I thought it'd be a good place to start. I've built the scene and started with some simple renders:

Perhaps I've taken the colour corrections a bit far and the vignetting is too much. However it's just fun to play around on something simple rather than overly grand ideas that never get finished. This initial rendering is just using Blinns and Lamberts. Nothing fancy yet. Single area light, ray traced shadows, and some very basic Final Gathering. I'm using an HDRI for the reflections though I can see that I need to boost the gain on them a bit as there's some crushing of the brights which is dropping them into the ugly grey area on the surface of the saucer.


Anyway, all that to say that I've also jumped on the 3D bandwagon. Hopefully all two of you that read this blog have a pair of red/blue 3D glasses. Strap yourselves in and hold on tight! You haven't seen anything this powerful since Avatar:

Small Update

Just posting the updated, rendered turntable. It still needs fixes, but I'm feeling a need to move on to something else. I'll probably still do the hair research and post updates of that as I figure things out.

Face Model Changes

We're on reading week now so since I finished my marking yesterday I thought I'd fix some of the most glaring issues with my face. Er... my CG face that is.

Firstly, since I rushed the texture painting process, I went back and painted out my double chin. This one isn't as obvious in the render, but combined with the geometric problems, it made me look fat. But I was still thin. But I looked fat. It was weird.

Next, using my favorite Sculpt Geometry tool in Maya, I went in and pushed my jaw and chin around, tweaked my lips, massaged around my eyes, and adjusted my ear ridges. Some of these modifications are more subtle than others, but in the end, I think this looks tons better. I still haven't done any real hair, but I've been playing with paint effects hair in Maya and perhaps I'll try and get a beard going first before I move on to real, styled hair.

As a last thing to look at, I've never been happy with the way blogger deals with videos, so I've gone back to my Face Off project and compiled a movie showing my progress because that is what I had originally built my face model for. So here's a breakdown of the five segments:

1- Previs. I needed a video to explain to the wife what she was filming and why. This also helps with the timing of when certain actions have to happen.

2- Proof of concept. Once it was filmed and I had it all tracked, I took my face model, put a chrome shader on it and loaded in the HDRI I took of my kitchen. It seemed to be good.

3- Next was a roto/CG test. I wanted to see that I could actually get my fingers looking like they were on top of the CG model of my face coming off. I used projection mapping to attach the texture to the face. I grabbed that right from the video footage.

4- After taking off the face, I wanted to see what it would look like with the hole that is left over. This is pretty rough, but started to give an idea of how much work was ahead. How much? Lots.

5- Lastly, I just wanted to see how it was all coming together so I grabbed a skull model I had, modified it, put a chrome shader on it and comped it all together. This really showed me just how hard this was going to be with the hand moving on top. If I was smart I would have filmed two versions (one without the hand motion) and I might have actually finished this project.

Just A Test

Here's the latest. I've modified some of the geometry and gave the background beyond the main canyons some focus. The sky image is from a sunset I shot last night out on garbage hill. If the weather holds up tonight I'll shoot a few more and pick the best, but this one seems to be working quite well for what I'm wanting. After roughing this together I can tell right now the biggest challenge will be to get the video footage to match the CG. Hopefully in the next week I'll be at the point where I can start detailing some of the canyons in Mudbox. I think the flatness of the geometry also isn't helping to blend the video footage and CG. On top of that, I think I need things like fallen boulders and whatever else one might find in desert (dessert) terrain (train).