Motion Retargeting via Spacetime Optimization of Rig Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion retargeting methods for character animation require manual intervention and editing, as they do not produce artist-friendly rig space parameters, leading to complex editing tasks due to high degrees of freedom and redundancy in rig controls, making it difficult to achieve desired poses efficiently.
Innovation Solution
A motion retargeting method that generates artist-friendly rig space parameters by creating an intermediate object from source motion and target character, extracting keyframes, and performing spacetime optimization to minimize pose errors, ensuring versatility, sparsity, and user control easiness, using an energy functional equation to optimize skeletal and deformation rig parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional motion retargeting methods are used to transfer motion data, then motion transfer can be achieved, but the resulting rig space parameters are not artist-friendly and require manual editing intervention
Solution Approach 1:
The patent introduces an intermediate object as a mediator between the source motion data and the target character rig. This intermediate object serves as a bridge that translates motion data into artist-friendly rig space parameters automatically, eliminating the need for manual editing while maintaining compatibility with the target character's control rig structure
Solution Approach 2:
The system automatically adjusts and transforms motion parameters from the source motion capture data into the specific parameter space required by the target character's rig. This parameter transformation process converts raw motion data into artist-friendly parameters that directly control the character's skeleton and deformation, making the results immediately usable without manual intervention
2Manufacturing precision
If all available rig parameters are utilized to achieve target pose, then pose accuracy can be improved, but the control becomes redundant and the editing task becomes very complex
Solution Approach 1:
The patent extracts and activates only the necessary subset of rig parameters required to achieve the target pose, rather than utilizing all available parameters. This selective activation removes redundant controls from the editing process, reducing complexity while maintaining pose accuracy by focusing only on the essential degree of freedom needed for the specific motion task
Solution Approach 2:
The system applies partial action by activating only the minimal necessary parameters rather than all available parameters. This approach achieves the desired pose accuracy using a subset of controls, avoiding the complexity that would arise from managing all redundant parameters simultaneously
3Adaptability or versatility
If motion retargeting produces results with high degrees of freedom and redundant controls, then versatility in motion processing is improved, but the editing task becomes very complex
Solution Approach 1:
The patent segments the rig control space into functional groups and activates only the relevant segments needed for the current motion task. This segmentation allows the system to maintain versatility by having access to all controls when needed, while simplifying editing by presenting only the necessary controls for the specific operation being performed
Data Source
AI summary
The present disclosure relates to a motion retargeting method and apparatus for character animation, and the method may include generating an intermediate object including at least one of a skeleton and a mesh based on a source motion and a target character, extracting keyframes of the intermediate object, constituting target rig parameters based on the extracted keyframes, and performing spacetime optimization on the target rig parameters to minimize a pose error between the intermediate object and the target character.