Before I start talking about the advanced rigging class, I would like to talk a little bit about the first time I met Chris Pagoria during my visit to Blue Sky Studios in March. Chris was with Sabine Heller when she came to pick me up from the train station in Connecticut. I had a great first impression of Chris as a very friendly, energetic, cooperative person, and of course an amazing rigger. Chris made sure that I was happy and doing well in New York City, he sent me so many recommendations of places to visit and he took me for a tour around downtown Manhattan. During our time together, I had the pleasure to chat with him about facial rigging, and this motivated me to enroll in his facial rigging class at AnimSchool.
Who’s Chris Pagoria?
Chris is a Character Technical Director at Blue Sky Studios. He’s one of the most talented facial riggers in the industry, he worked on Epic (2013), and was the character rigger of Mary Katherine, one of my favorite CG characters. He also worked on Ice Age 4: Continental Drift, which was his first movie with Blue Sky Studios. Previously, he worked at Rhythm & Hues and spent two years in television animation. Chris is a super-fast modeler, and one of Chris’s best talents is that he can model facial blend shapes very fast, he has an amazing sense of art direction and an eye for very small details. I liked his way of teaching facial rigging; he simplifies everything and he is very direct to the point.
Advanced Rigging Class
This class is about character facial rigging, with a focus on modeling facial blend shapes. The main goal is to rig a complete head from scratch! The first few weeks in the class are focused on creating mouth shapes and appealing smiles. After that we moved on to the eyebrows shapes, the eyes rig, and finally the head deformations and the rig final touches.
The Main Mouth Shapes and their Combos!
The main mouth shapes are the smile and the frown, which can be wide or narrow. They are fired by the mouth corner control using Translate X and Translate Y. In my opinion, these shapes are the most important shapes in the facial rig, and you have to spend a good amount of time working on them to make them look appealing and function well.
The diagram below shows the mouth corner control function when you move it on the X-Y plane. First of all you sculpt the main four shapes (Smile, Frown, Wide and Narrow), then you combine each two shapes to get the basic combo shape, then you modify it and connect it to the rig. We connect the combos to the rig using render nodes that tells Maya to fire this corrected shape if the two shapes get fired.
I wrote a python script for creating and editing these combos on the spot, please check the following article: http://eyad.tv/home/easycombo-script/
More Mouth Shapes
What I learnt in this class is that you have to make sure that all of the face shapes work together smoothly. To do that, always think about how the shapes are working together; for example, if you want to sculpt a pucker shape, you can fire 30% narrow, and then from that shape you start sculpting the pucker. If you use this method, you will get nice motion flow between the shapes, and they will look very consistent. The image below shows narrow and pucker shapes running at the same time and they work smoothly together without any corrective.
The other shapes that are important in the mouth rig are the sneers, the lip roll, the lip press, and the lip puff. I will just talk about the sneers here, and how they are created. The idea of the sneers is that you create two main shapes, “All Lips Up” and “All Lips Down”, then you split each shape into upper lip and lower lip, and these shapes you split them again into three parts, left, mid and right.
The jaw is divided into two sections, the artistic side and the technical side. The idea is that you create a joint on each span of the mouth and these joints will be driven by punch of joints that are located in the mouth jaw. Those punches of jaw joints will help you later to do the mouth zipper and the mouth corners sharper.
After you are finished from painting the skin weights for the jaw opening, you have to create some correctives for the jaw shape. The image below shows the jaw rotated 20 degrees, and fired a corrective that fixes the jaw line and the area between jaw and neck.
One of the main problems that you will face with the mouth opening is that you lose the sharpness of the smile and of the frown. Chris taught us a cool setup that fix that problem, so when you open the mouth, then do a smile, the mouth corner joints will move automatically and sharpen the corner of the mouth.
The eyebrows are very fun to do, and you can do them exactly like the mouth sneers. There are two main shapes, the “Brows Up” and “Brows Down”, then you split them into In, Mid and Out. When sculpting these two shapes, the main idea is to keep the brows sliding on the skull, which will make your shapes very clean and interesting.
The tricky part with splitting the brows shapes is having them work smoothly with each other so that they create nice arches. There are also some other shapes for the brows, like the squeeze and the brows rotation.
There are several ways to rig the eyelids; my favorite is the one that combines the technical and the artistic features in one rig. The idea behind Jane’s eyelids rig is that there is a joint on each span of the lids, almost like the mouth rig, and their parents are in the middle of the eyeball, so when the parents rotate the eyelid’s joints they will move on the spherical surface of the eye. I created a locator on each joint and aim constrained the parent joints to them, so when you move the locators in the space, they will drive the rotation of the parents joints, therefore they will shape the eyelid. Then I built some layers (groups) over these locators, and on each layer I put a specific function, for example in “Layer 1”, upper lid closing, “Layer 2”, upper lid rotating, and so on. I used many different methods to drive these locators, like for example set driven keys, expressions, render nodes, etc.
When you are finished with shaping the locators, the artistic side of creating the eyelids comes into play. Now you can make the eye corrective shapes, so that the eye looks more appealing and responds well to specific emotions.
The nose rig is very simple, made by only one joint in the middle, plus two shapes for the nose sneer and flare. You can go a little bit more advanced if you like, by combining the sneer and the flare in a corrective.
Using specific controls you can deform the head to create squash and stretch behavior. These controls are combinations of lattices and blend shapes. I always like to divide the head into two parts, upper head and lower head. These two joints are located under the horizon of the lower eyelid, and you can also use them to tilt the head.
I wrote a script that generates combos automatically, and I used it to do the final touches on the rig.
For more information about the script, please visit the following link: http://www.eyad.tv/blog/?p=722/
Chris asked us to make five facial expressions to show the final result of the rig. I experimented with many different facial expressions and I chose ten of them to present in the motion test video.
I’d like to thank Chris Pagoria for his amazing efforts in this class, my classmates, and the AnimSchool Family.
Also, a special thank you to Dave Gallagher, the founder of AnimSchool, who gave me a very long and detailed review on Jane’s facial rig. He showed me how to pose the face right, and how to make appealing shapes for the mouth. It was one of the best and most helpful reviews I’ve ever had.
Feel free to download the wallpapers below!