Motion Sequence Determination Using Sub-Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion matching techniques face difficulties in handling various contexts due to the need to change the entire feature vector when the game situation changes, making them inflexible.
Innovation Solution
A system that includes a motion database, a sub-feature extraction unit, and a motion-sequence determination unit, which extracts and compares sub-features related to the virtual object's state to determine a suitable motion sequence from the database, allowing for more flexible motion matching by enabling context-dependent sub-feature extraction and similarity calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional motion matching uses the whole feature vector for motion sequence determination, then the motion matching can handle various contexts, but the system complexity and computational burden increase when context changes
Solution Approach 1:
The patent segments the feature vector into multiple sub-feature vectors, where each sub-feature vector corresponds to a specific type of feature (e.g., spatial, temporal, motion dynamics). This segmentation allows the system to selectively process only relevant sub-features for each motion matching task, reducing computational complexity while maintaining adaptability to different contexts.
Solution Approach 2:
The patent implements dynamic selection of sub-features based on the current game context or situation. The system can adaptively choose which sub-feature vectors to extract and compare depending on the specific motion matching requirements, allowing flexibility without requiring the entire feature vector to be processed in every case.
2Measurement precision
If conventional motion matching extracts all features from the virtual object, then comprehensive motion analysis is achieved, but the processing time and computational resources increase
Solution Approach 1:
The patent extracts only the necessary sub-features from the complete feature vector based on the specific motion matching task. Instead of processing all features uniformly, the system selectively extracts relevant sub-feature vectors (e.g., only spatial features for position-based matching, or only temporal features for timing-based matching), thereby reducing processing time while maintaining analysis precision for the relevant aspects.
Solution Approach 2:
The patent applies partial action by processing only a subset of features (sub-features) rather than the complete feature vector. This partial processing approach is sufficient for many motion matching scenarios where not all features are equally relevant, thus achieving acceptable precision with reduced computational overhead and faster processing.
3Manufacturing precision
If conventional motion matching changes the whole feature vector when context changes, then accurate motion matching for different situations is achieved, but the flexibility and ease of operation decrease
Solution Approach 1:
By segmenting the feature vector into independent sub-feature vectors, the patent allows the system to maintain consistent sub-feature extraction methods across different contexts while only changing the selection or weighting of specific sub-features. This segmentation makes the system easier to operate and adapt, as developers can modify individual sub-feature handling without redesigning the entire feature processing pipeline.
Solution Approach 2:
The patent creates a universal sub-feature extraction framework that can handle multiple contexts and game situations through a common architecture. The same sub-feature extraction unit can serve different motion matching scenarios by selectively processing different sub-feature types, providing both accuracy across contexts and ease of operation through a unified, flexible system.
Data Source
AI summary
One or more embodiments of the invention is a system for determining a motion sequence of a virtual object, the system including: a motion database that stores each of a plurality of motion sequences in association with sub-features of a plurality of types, each of the sub-features of the plurality of types associated with said each motion sequence being a feature related to said motion sequence; a sub-feature extraction unit that extracts a sub-feature(s) of a predetermined type(s) from features related to the virtual object; and a motion-sequence determination unit that determines one motion sequence from the motion database by comparing each of the sub-feature(s) of the predetermined type(s) extracted by the sub-feature extraction unit with each of sub-feature(s) of the predetermined type(s) in the individual motion sequences stored in the motion database.