Blended Animation System for Coarse and Fine Motion Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning systems are capable of generating coarse joint positions for animating virtual agents but struggle with fine-grained movements such as facial expressions, which are typically pre-authored or manually crafted.
Innovation Solution
A computing system that blends motion outputs from a machine learning system with manually crafted motion outputs on a frame-by-frame basis, using a motion controller to generate a motion input vector, a motion generator to produce a motion output vector and pose information, selecting an animated motion from a bank that matches the pose information, and applying blending coefficients to generate a blended animation, while also providing a reward signal for animation quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If machine learning systems are used to generate animation, then productivity is improved through automated motion generation, but manufacturing precision deteriorates due to inability to produce fine-grained movements
Solution Approach 1:
The patent combines machine learning-generated coarse motions with manually crafted fine-grained motions into a unified animation output. The system merges two different motion generation approaches (automated ML and manual authoring) to simultaneously achieve high productivity from ML and high precision from manual crafting, resolving the contradiction between automation efficiency and animation quality
Solution Approach 2:
The animation generation process is segmented into two distinct components: coarse joint positions generated by machine learning and fine-grained movements manually crafted. This segmentation allows each component to specialize - ML handles overall motion efficiency while manual authoring handles precision details - thereby resolving the contradiction between productivity and precision
2Manufacturing precision
If manually crafted animations are used for fine-grained movements, then manufacturing precision is improved, but device complexity increases due to multiple motion sources
Solution Approach 1:
The patent introduces a motion blending module as an intermediary that automatically combines coarse and fine-grained motions. This mediator handles the complexity of integrating multiple motion sources, allowing the system to achieve high precision facial expressions while the blending module manages the computational complexity of coordinating multiple motion layers
3Manufacturing precision
If pre-authored motions are used for fine-grained movements, then manufacturing precision is improved, but adaptability deteriorates due to limited motion variations
Solution Approach 1:
The system makes the animation adaptive by dynamically selecting and blending pre-authored motions with ML-generated variations. The motion blending module adjusts the combination of pre-authored and generated motions based on contextual requirements, allowing the system to maintain high precision from pre-authored motions while achieving adaptability through dynamic motion selection and generation
Data Source
AI summary
In one implementation, a method for generating a blended animation. The method includes: obtaining a motion input vector for a current time period; generating a motion output vector and pose information for the current time period based on the motion input vector; selecting an animated motion from a bank of animated motions for the current time period that matches the pose information within a threshold tolerance value; obtaining a blending coefficients vector for the current time period; generating a blended animation for the current time period by blending the motion output vector with the animated motion based on the blending coefficients vector; and generating a reward signal for the blended animation for the current time period.