MotionCanvas generates full-body motion from diverse kinematic inputs
MotionCanvas: Learning Implicit Motion Planning from Composable Kinematic Cues
Multimedia
Summary
Creating realistic character movements in animation is tricky because animators want to control exact details, like arm swings or walking paths. The authors designed MotionCanvas, a tool that takes various motion clues—like joint positions, rotations, and even language instructions—and smoothly combines them to create natural and precise full-body animations. It learns from many examples with different combinations of clues to fill in missing details and produce coherent movements. This approach works well for tasks like editing animations by instructions, fixing motion glitches, and generating sequences step-by-step.
What this means in practice
- •For animation artists: Create precise yet natural full-body animations by combining diverse motion constraints and language inputs in one seamless process.
- •For game developers: Generate or edit character movement sequences on the fly using specified body part motions and instructions to improve in-game animation control.
Authors
Zeyu Ling, Di Kang, Qing Shuai, Yuxin Wen, Jing Li, Zhanke Wang, Heng Li, Changqing Zou, Chunchao Guo, Linchao Bao
Abstract
Professional character animation requires both natural motion and precise, versatile control. For example, it is common for the creators to define the timing of a specified action, to control the motion range of the character's arm swing, and the route the character walks through, like specifying various kinematic motion cues on a ``motion canvas''. This motivates us to propose MotionCanvas, a model that supports \emph{cue-conditioned implicit motion planning} to faithfully and coherently connect all cues, dense or sparse, full or partial, into one full-body motion sequence. Specifically, MotionCanvas represents heterogeneous kinematic cues on a shared motion canvas, where position and rotation values are specified across body joints and time. A shared flow-matching model generates motion conditioned on this canvas, with optional language and input motion; cue imputation keeps the specified canvas values fixed in both training and sampling. To learn coherent completion across different cue sets, we train with a compositional cue sampler that varies when cues are applied, which positions or rotations are specified, and how they are combined. Together, these designs enable a single generator to synthesize globally coherent actions that jointly satisfy compatible heterogeneous cues. We test this planning ability with temporal, root, and body-part cues---alone and in combination---and language-guided editing. We naturally extend this evaluation to sequential generation and motion repair, since both require the same ability to organize coherent motion from kinematic cues. Across these evaluations, MotionCanvas establishes state-of-the-art results in controlled-motion quality, mixed-cue adherence, sequential generation, instruction editing, and motion repair while preserving its text-to-motion capability.