3D Volume Recognition via Sub-Cluster Centroid Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D imaging systems for gesture recognition lack the ability to accurately recognize volumes, leading to limited and inaccurate interaction with data processing systems, especially in applications requiring finer volume recognition.
Innovation Solution
A volume recognition method that groups points in a 3D imaging system into sub-clusters based on position and size, associating a volume with each sub-cluster's centroid, allowing for a detailed three-dimensional model of objects, which can be processed efficiently for interaction with data processing systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all objects in the image are allocated a blob for volume recognition, then the recognition accuracy of object volumes is improved, but the data processing capabilities are exceeded and the system becomes impractical for fine volume recognition
Solution Approach 1:
The patent segments the set of all detected objects into a subset of 'objects of interest' based on relevance criteria. Instead of processing all objects uniformly, the system identifies and focuses computational resources on specific objects that meet predetermined interest criteria, thereby reducing the total number of volumes to be recognized while maintaining accuracy for critical objects.
Solution Approach 2:
The patent applies different processing quality levels to different objects based on their interest status. Objects of interest receive fine volume recognition with detailed sub-clustering, while non-interest objects receive coarser processing or are excluded. This local differentiation optimizes processing power allocation to match actual system needs.
2Productivity
If the points of the 3D image are grouped in clusters according to perceived depth, then the number of objects to be processed is reduced, but the volume recognition becomes too crude for applications requiring finer detail such as gesture recognition
Solution Approach 1:
The patent performs a second-level segmentation by dividing clusters into sub-clusters. After initial clustering groups points by depth, the system further subdivides relevant clusters into sub-clusters that represent finer volumetric structures. This hierarchical segmentation enables detailed volume recognition for objects of interest while maintaining processing efficiency through the initial clustering step.
Solution Approach 2:
The patent implements a nested structure where sub-clusters are contained within clusters, which are themselves contained within the overall 3D image data. This nested organization allows the system to work with coarse cluster representations for general processing while enabling fine-grained sub-cluster analysis when detailed volume recognition is required for specific objects.
3Measurement precision
If a large number of sub-clusters are created for detailed volume recognition, then the accuracy of three-dimensional modeling is improved, but the dataset size and processing requirements increase significantly
Solution Approach 1:
The patent creates detailed sub-clusters only for objects of interest rather than uniformly across all detected objects. By applying fine-grained volumetric segmentation locally to specific target objects while using coarser representation elsewhere, the system achieves high modeling accuracy where needed while keeping the overall dataset size manageable.
Data Source
AI summary
The present invention relates to a volume recognition method comprising the steps of:a) capturing three-dimensional image data using a 3D imaging system 3, wherein said image data represent a plurality of points 5, each point 5 having at least a set of coordinates in a three-dimensional space;b) grouping at least some of the points 5 in a set of clusters 6; c) selecting, according to a first set of parameters such as position and size, a cluster 6 corresponding to an object of interest 1 located in range of said imaging system 3; d) grouping at least some of the points 5 of the selected cluster 6 in a set of sub-clusters according to a second set of parameters comprising their positions in the three-dimensional space, wherein each sub-cluster has a centroid 11 in the three-dimensional space; ande) associating a volume 12 to each of at least some of said sub-clusters, wherein said volume 12 is fixed to the centroid 11 of said sub-cluster. The present invention also relates to a volume recognition system for carrying out this method.