Wrist-Mounted Hand Pose Tracking via Distributed Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing hand gesture recognition technologies face challenges in capturing and recognizing continuous hand poses and fine-grained finger gestures, especially when the hand moves outside the camera's field of view or requires obtrusive wearable devices.
Innovation Solution
A non-obtrusive wrist-mounted system with multiple imaging sensors, closely spaced to the skin, captures images from anatomically distal positions and uses machine learning algorithms to determine hand poses and finger positions, including those not directly in the field of view, through stitching and kinematic modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional computer vision cameras are used to capture hand poses, then hand gesture recognition is possible, but the system fails when the hand moves outside the camera's field of view
Solution Approach 1:
The patent segments the imaging function across multiple sensors positioned at different locations on the wearable device. This allows different sensors to capture different portions of the hand or different views, ensuring continuous tracking even when the hand moves outside any single sensor's field of view.
Solution Approach 2:
The patent transitions from a single fixed camera viewpoint to a distributed array of imaging sensors on a wearable device. This spatial redistribution creates multiple viewing dimensions, allowing the system to track hand poses from various angles and maintain reliability as the hand moves through three-dimensional space.
2Duration of action of moving object
If wearable gloves or position sensors are used to estimate hand poses, then continuous hand tracking is enabled, but the devices become obtrusive and only recognize limited discrete poses
Solution Approach 1:
The patent replaces mechanical sensors (gloves with position sensors or markers) with optical imaging sensors. This substitution eliminates the need for physical contact with the hand, making the system non-obtrusive while enabling continuous visual tracking of natural hand movements and gestures.
Solution Approach 2:
The imaging sensor system serves multiple functions: it captures continuous video for tracking hand position, recognizes discrete gestures, and provides fine-grained finger pose estimation. This multi-functionality replaces the limited discrete pose recognition of wearable sensors with comprehensive continuous monitoring.
3Ease of operation
If imaging sensors are placed far from the skin to avoid obstruction, then device comfort is improved, but measurement accuracy of hand poses decreases
Solution Approach 1:
The patent employs a flexible wearable band structure that allows imaging sensors to be positioned in close proximity to the skin without causing discomfort or obstruction. The thin, flexible form factor enables the sensors to conform to the wrist anatomy, maintaining both comfort and measurement precision.
Solution Approach 2:
The patent uses multiple imaging sensors to capture multiple views of the hand, creating redundant visual information. This allows the system to accurately reconstruct hand poses even when individual sensors are positioned close to the skin, as the combined data from multiple sensors compensates for any single sensor's limited perspective.
Data Source
AI summary
A wrist-mounted system for tracking hand poses includes one or more cameras mounted to a wearable band. In some examples, the one or more cameras are low profile and may be located less than 15 mm, and preferably less than 10 mm, of the wrist of the user. In some examples, a system includes a wearable band and one or more imaging sensors. The one or more imaging sensors are disposed to have a field of view that is anatomically distal when the wearable band is coupled to an arm of the user. The one or more imaging sensors each define an optical axis spaced from the wearable band of less than 15 mm, and preferably less than 10 mm. The image data may come from a single camera mounted on the back of the wrist which captures images of the surface contours of the back of the hand to infer hand poses and gestures.