Fingertip Centroid Identification via Occupancy Blob Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional 3D or depth sensing camera systems face challenges in accurately detecting and tracking fingertips, leading to inaccuracy and instability, especially when only a part of the hand is visible, and are unsuitable for precise finger-to-item interactions.
Innovation Solution
A computer-implemented method using kernels to identify fingertips by analyzing image data from a camera system, calculating centroids for each identified fingertip, and tracking these centroids in real-time, allowing for accurate detection and tracking even when parts of the hand are not visible, and enabling precise finger interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If skeletal tracking with inverse kinematics is used, then hand pose can be estimated, but fingertip position accuracy deteriorates because joints are repositioned off-centre
Solution Approach 1:
The patent extracts fingertip detection from the skeletal tracking system by using independent 2D blob analysis on hand region images. This separates the fingertip localization task from the joint-based skeletal model, allowing accurate fingertip detection without being constrained by inverse kinematics joint repositioning errors.
Solution Approach 2:
The patent introduces 2D occupancy blob analysis as an intermediary step between raw depth data and fingertip position. By converting depth data to 2D occupancy maps and analyzing blob characteristics, the system achieves accurate fingertip localization without directly relying on skeletal joint estimates.
2Reliability
If hand centre tracking is used, then cursor positioning is simplified, but tracking reliability deteriorates when hand shape changes or hand is partially in view
Solution Approach 1:
The patent segments the hand into individual finger regions and detects fingertips independently rather than treating the hand as a single blob. This segmentation allows accurate fingertip tracking even when the hand is partially in view or when fingers are curled, maintaining reliability while preserving precise cursor control.
Solution Approach 2:
The patent applies different analysis methods to different parts of the hand - using 2D occupancy blob analysis specifically for fingertip regions rather than uniform hand centre tracking. This local quality approach maintains tracking stability for visible fingertips while enabling precise cursor control through accurate fingertip position detection.
3Measurement precision
If conventional depth sensing is used, then 3D hand data can be captured, but fingertip detection accuracy deteriorates due to inverse kinematics joint repositioning
Solution Approach 1:
The patent replaces the mechanical inverse kinematics joint-based system with a 2D image processing approach using occupancy blobs. By substituting the 3D skeletal model with 2D blob analysis on hand region images, the system achieves accurate fingertip detection with reduced processing complexity.
Data Source
AI summary
A computer implemented method of fingertip centroid identification in real-time, implemented on a computer system comprising a processor, memory, and a camera system. The processor receives image data from the camera system; runs a first kernel comprising a set of concentric closed shapes over image data to identify an occupancy pattern in which internal closed shapes are at least nearly fully occupied, and in which a subsequent closed shape has at least a relatively low occupancy level, so as to identify one or more fingertips in the image data; for each identified fingertip, runs a second kernel over the identified one or more fingertips to establish a best fit closed shape which covers each identified fingertip; calculates a centroid for each best fit closed shape which corresponds to an identified fingertip; and stores in the memory the calculated centroids for the identified one or more fingertips.