Semantic Labeling of Negative Spaces in 3D Scene Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for generating and utilizing three-dimensional scene models in extended reality (XR) environments face challenges in accurately representing and interacting with physical environments, particularly in defining and utilizing negative spaces with semantic labels, which are crucial for realistic and context-aware XR experiences.
Innovation Solution
The method involves obtaining a three-dimensional scene model of a physical environment by spatially disambiguating points into clusters and subspaces, assigning semantic labels to these clusters and subspaces, and using characterization vectors to determine the spatial extent and semantic labels of subspaces, allowing for the generation and display of XR representations that navigate and interact with the environment based on these labels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If point clouds are used to represent physical environments in XR applications, then three-dimensional spatial representation is achieved, but negative spaces (voids between objects) cannot be accurately defined or utilized
Solution Approach 1:
The patent segments the three-dimensional space by dividing it into multiple subspaces, each potentially containing points or representing negative spaces. This segmentation allows the system to individually identify and label negative spaces between objects, enabling accurate representation of voids while maintaining overall spatial accuracy.
Solution Approach 2:
The patent introduces characterization vectors as an intermediary data structure that bridges the gap between raw point cloud data and semantic understanding of negative spaces. These vectors encode spatial relationships and properties, enabling the system to infer and utilize negative spaces that are not directly represented by points.
2Adaptability or versatility
If semantic labels are assigned to all subspaces including negative spaces, then context-aware XR interactions are enabled, but computational complexity and processing time increase
Solution Approach 1:
The patent applies local quality by assigning semantic labels selectively to specific subspaces based on their characteristics. Rather than uniformly processing all subspaces, the system identifies negative spaces with specific properties (e.g., enclosed vs. open) and applies appropriate semantic labels only where needed, reducing overall processing complexity while maintaining context-aware capabilities.
Solution Approach 2:
The patent implements partial action by focusing computational resources on identifying and labeling only the most relevant negative spaces for XR interactions. Instead of exhaustively processing every possible subspace, the system prioritizes negative spaces that have significant impact on user interaction, thereby reducing processing time while preserving essential context-aware functionality.
3Reliability
If three-dimensional scene models include detailed negative space definitions, then XR immersion and realism are enhanced, but data storage requirements and transmission bandwidth increase
Solution Approach 1:
The patent uses characterization vectors as compressed representations or copies of the essential properties of negative spaces. Rather than storing complete geometric definitions of all negative spaces, the system stores compact characterization vectors that capture the critical spatial and semantic information needed for XR immersion, significantly reducing data volume while preserving immersion quality.
Data Source
AI summary
In one implementation, a method of defining a negative space in a three-dimensional scene model is performed at a device including a processor and non-transitory memory. The method includes obtaining a three-dimensional scene model of a physical environment including a plurality of points, wherein each of the plurality of points is associated with a set of coordinates in a three-dimensional space. The method includes defining a subspace in the three-dimensional space with less than a threshold number of the plurality of points. The method includes determining a semantic label for the subspace. The method includes generating a characterization vector of the subspace, wherein the characterization vector includes the spatial extent of the subspace and the semantic label.