Previs and postvis
Previs, or previsualization, is the visualization of a complex scene before filming/animating. It is used alongside storyboarding, working out every single shot in a complicated scene. Postvis is very similar to previs. It is used in sequences that are full of VFX. Editors/animators will use placeholders, or blank slugs, of CG characters and other elements layered over the edited footage until the rendering is finalised.
VFX vs Film vs Games
The animator’s role does in fact differ in these three fields, in some cases the differences are major ones and in other cases they don’t differ too much. There are pros and cons to working in either one of these three fields but these characteristics do overlap very often.
In games the character and the camera is in the gamer’s complete control, so the animation needs to look perfect at every possible angle. The animator needs to take into consideration whether the game is in first person, or third person. The animator needs to make sure the walk or run cycle of a certain character is following a smooth arc from any limb and seen from any angle. However, walk/run cycles only need to be animated once as they loop forever in games whereas in film an entire run sequence needs to be animated step-by-step in full. Characters in games usually undergo a lot of body mechanics, so an animator wanting to work in the games industry will have to perfect his body mechanics skills. Performance animation is still important as games are becoming more and more story-driven and require much of the same preparation as films do (research, previs, concept art etc.) but games are usually body mechanics-heavy. The turnaround for an animation in games is generally much shorter than film.
In film (and by film I mean Pixar or Disney) the animator only needs to worry about how the animation will look from one camera angle at a time. This allows the animator to ‘cheat’ or cut corners when animating. Ultimately if it looks perfect from that angle, then it’s OK. An animator working in film will be versatile in his skillset. Animators in film can do walk/run cycles, body mechanics, phonemes, performances and acting. Animators are expected to work on a single shot for weeks, even months at a time. Usually films are much less realistic-looking than games (just compare Toy Story to GTA5) but both industries rely on motion capture to make a character’s movements as realistic as possible, and many more studios and companies are implementing mocap into their pipelines. It makes the whole process a lot faster.
Props animation, particle systems.
Props in animation are objects that are stuck to a characters hand or body to be used in certain animations. Props are widely used in film, TV, VFX and games.
Particle systems are techniques used in games, films, VFX and motion graphics that use many 3D models or other CG objects that emulate ‘fuzzy’ phenomena such as chaotic systems, natural phenomena, explosions, fire, smoke or effects caused by chemical reactions – effects which are usually very difficult and time-consuming to reproduce with conventional rendering techniques.
Rotomation
Rotomation is basically rotoscoping. It’s a term used for tracing over a certain movement. It was originally used to trace over live action footage of actors frame by frame to replicate the movement of the actors in animation. However, this techniques was abandoned for a while as the animation wasn’t believable enough. The characters would appear to move and walk weightlessly. Nowadays rotomation is used in VFX, film and TV to separate an element from the background so another element can be placed between the two. Studios use this technique very often, most big budget, VFX heavy films use this technique.
Motion capture, matchmove, rotomation
Motion capture, or mocap, is the process of recording movement of objects or people. It is used in many different industries outside of film, VFX and games. The mocap recordings are turned into data which will then be matched to animated CG characters, both in 2D or 3D.
Matchmove, or motion tracking, is the process which allows computer graphics to be imported into live-action footage correctly. It is often confused with motion capture, but it does not capture the movement of individual objects or people, but it emulates the movement of the camera itself and recreates its path digitally – it’s essentially creating a virtual map of the shot and an identical virtual camera movement.
Rotomation is very similar to rotoscoping, but with a slightly different purpose. Rotomation is tracing over a certain movement so it can be separated from the background so a different CG element/object/effect can be placed in the middle, between the background and the traced movement in the foreground.
Luca's research blog 