On the beauty of things
Rigging
My rigging process always starts with the analysis of the characters, the assets and the animation style needs and particularities.
In Shim’s case, its character design and personality needed me to concentrate more on the body mechanic aspect than the facial rig. So I started with a basic biped rig, and I then installed an IK/FK switch on both legs and arms. Once this first step was done I added attributes to match the specificities of the character. Shim being quite short and stubby, we needed a way to make animating him easier, so I built an automatic stretch in the legs and the spine.[a] Another particularity of his rig is texture rigging. At some point in the film, Shim’s hands go from a textured green to a smooth white passing by a textured black. To rig this transition I worked with Noée who created a ramp which I connected to controllers. Once the rig for the body was built, I started skinning.[b][c][d]
After the body rigging was done and Noa was able to start animating, I went on to rigging the poncho. At first, we wanted to use an auto-collision similar to a skirt rig, where when the legs move, they move the part of the poncho they are in contact with with them.
However, the poncho was too short for this type of rig to work. I went back to the drawing board and decided on a more classic rig for loose clothing. I placed chains of joints all around the collar, kind of resembling an umbrella’s ribs, and simply connected controllers to each joint allowing the poncho to be moved all around the collar and the hem but also in the middle of the fabric and along the arms.
For the hat, it was basically the same as for the poncho. I just added a root controller so that when animating, we would be able to take it off Shim’s head and constrain it to the hands.
I finished the rig with facial rigging, which was very simplified thanks to the hat hiding half of the face. I used clusters that I converted to blendshapes to rig the mouth and the lips[e][f].
For most of the moulds I went for the simple route and used chains of joints, each of them connected to a controller so we could animate them swinging.
[a] "So I started with a basic biped rig, [I placed a 3d object named a joint everywhere an articulation is needed in the skeleton, then connected a controller for each joint.]" ;
[b] controllers [These are standard geometric shapes commonly found in most software. For example, the diamond is used for rotation and is connected to the hands or feet. The trapezoid is used for joints and is generally located at the knees or elbows. As for the sphere, it’s a controller typically used at the neck, head, or hips.] ;
[c] skinning [This is indeed the method used to attach the character's skin to its skeleton.] ;
[d] IK/FK switch [IK : "Inverse Kinematics", FK : "Forward Kinematics"] ;
[e] clusters [Clusters are a type of deformer commonly used in the rigging process. This particular deformer allows for control over the points of an object, such as the vertices of a polygon or the CVs of a curve.] ;
[f] blendshapes [It's a feature found in every 3D software that allows bones to be linked to polygons. To do this, you paint a zone of influence on the geometry for each bone.]
At first excitement ? I actually really enjoy rigging so I couldn’t wait to start this part of the project. All was fun until stress piled up and I started making small mistakes. To this was added bugs and whatnot so I inevitably fell behind on tasks.
Looking back I learnt so much both purely in rigging and in problem solving and in the end the rig worked well and was easy to animate so I am now less hard on myself and I am very proud. But during crunch time, I think I almost started praying lol.
Shim’s character design is really endearing, so funny enough it made the fact that it was on my screen for months more bearable.
The rigging itself being very straightforward was also something I liked about it. Compared to the moulds that I had to really polish to avoid harsh angles, Shim’s rig was really simple (not in an easy way more like plain ? But positive), just like his personality and design ! This made it easier for me to determine what I had to do in what order without getting lost or overwhelmed (at first at least).
The definitive negative part of Shim’s rig was the poncho. Which was actually really not complicated, but I was stuck on this first idea that we had and didn’t want to let it go for some reason, even if it didn’t work at all. So I lost time and energy on this and felt desperate until I finally let it rest for a while and came back to it, I think a whole month later in the middle of animating, and remade the whole thing successfully in one evening.
For the moulds everything went smoothly. It was fun to adapt this very simple rig to the different shapes and needs depending on the mould variety. Do they jiggle ? How tall are they ? Will they be in the storm ? Does Shim interact with them ? Apart from the polishing mentioned earlier that was really the only thing a little troublesome, they were very nice to work on !
Acting & Animation
I like to imagine the movement first. I focus on the way I would decompose the action, how the weight balances, how the environment influences the scene, etc… Shim is a round little guy, it’s moving like a squishy marshmallow. Despite its body shape, it is very delicate and soft so I had to consider that factor for the animation. We had filmed a lot of references already so it helped with the visualisation. Analyzing the reference videos is very important to deconstruct the moves and understanding them. It makes the animation process faster. The main goal is to make sure the poses of the body are organic and accurate, while keeping in mind the way Shim moves is altered by his mushroom-like nature.
I feel very motherly towards Shim. Giving life to its body was such an emotional process, especially towards the end of the production, when exhaustion was the strongest. I was worried all year long about “what if my skills are not good enough,” “what if I can’t animate Shim to its full potential.” Being emotionally attached to the characters you create makes you feel like you also have the possibility to mess them up if you don’t treat them correctly (life lesson for real). I would spiral a lot, and then worry it was slowing me down so I’d push my thoughts aside to complete my work, and repeat. Now it’s done and I’m picking up all the experience and lessons this animation work gave me. Seeing Shim move and evolve in its environment and garden its molds made a lot of our team’s hard work shine so bright and I’m very proud of our accomplishments.
Expression
We decided very early on that the focus of our movie was the gardens, so we agreed on a more discreet character design for Shim. His eyes are hidden behind his triangle hat, only the lower half of his face is visible. A lot of the time, facial expressions were not central in the scenes. That said, I used my own face as direct reference, making faces to understand how the muscles around my mouth and nose were working. Sometimes I would remember a precise scene from a movie I watched that would match the emotion I was trying to portray.
Scenes where the action was heavy were the funniest scenes to animate. Non cycling movements are usually the easiest for me, as I feel I have a lot more freedom about the way they carry out. The most common inconveniences I’d met would lie in technicality, like ease-in and ease-out for every movement or making sure trajectories are clean, though it was still manageable. Shim was well-behaved most of the time as Leo, our rig artist, did an amazing job that facilitated the animation (its poncho was the biggest problem, acting like it had a will on its own). However, walk and run cycles were really upsetting. I couldn’t for the life of me get it theory-right. Our teacher Paul and I would usually get very distressed about how bad I was in run cycles. The team found them acceptable (thankfully) but I was so upset that I couldn’t get it right.