Multi-Pointer Gesture Recognition via Touch Point Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Interactive input systems employing two imaging devices face challenges in accurately locating and tracking multiple pointers, leading to pointer ambiguity issues, especially when the pointers do not have distinctive markings, which complicates the recognition of gestures made using multiple pointers.
Innovation Solution
A gesture recognition method that captures images across an input region, processes them to identify clusters of touch points, recognizes gestures based on the motion of these clusters, and updates the display accordingly, using at least one imaging sensor and processing structure to analyze image data and determine the location of touch points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple pointers are tracked using triangulation with two imaging devices, then pointer location can be determined, but pointer ambiguity issues occur when pointers lack distinctive markings
Solution Approach 1:
The patent segments the set of all detected touch points into multiple clusters, where each cluster corresponds to a single pointer. This is achieved by analyzing spatial relationships and grouping touch points that belong together based on their relative positions and motion patterns, thereby resolving the ambiguity of which touch points belong to which pointer
Solution Approach 2:
The patent introduces cluster centers as intermediary representations that simplify pointer identification. Instead of directly tracking individual touch points which may be ambiguous, the system computes cluster centers from grouped touch points and tracks these centers, providing a stable intermediary representation that resolves pointer ambiguity
2Measurement precision
If cluster analysis is performed to resolve pointer ambiguity, then gesture recognition accuracy improves, but processing complexity increases
Solution Approach 1:
The patent divides the processing task into distinct stages: first detecting all touch points, then segmenting them into clusters based on spatial relationships, computing cluster centers, and finally tracking gesture motion. This segmentation of processing steps makes the complex task more manageable and efficient
Solution Approach 2:
The patent applies clustering and center computation only to the necessary subset of touch points that form actual pointers, rather than processing all detected points equally. This partial action approach reduces unnecessary computation while maintaining gesture recognition accuracy
Data Source
AI summary
A gesture recognition method comprises capturing images, processing the images to identify at least two clusters of touch points associated with at least two pointers, recognizing a gesture based on motion of the clusters, and updating a display in accordance with the recognized gesture.