Single Camera Gesture Recognition via Reflector Occlusion Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional touch systems struggle to accurately determine the positions of multiple touch points due to occlusion, leading to incorrect gesture recognition and increased system costs with multiple cameras.
Innovation Solution
An interactive system using a light source, reflector, and image sensor captures image windows with object images, allowing for gesture recognition by analyzing the relations between object images, such as average number and distance, to perform gestures like rotation, zoom, and scrolling without calculating exact touch point coordinates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple cameras are arranged at different corners to capture image windows, then the accuracy of determining multiple touch points is improved, but the system cost increases
Solution Approach 1:
The patent uses a single camera to capture both the real object images of fingers and their virtual images formed by the reflector. The reflector creates virtual copies of the light paths, allowing the single camera to effectively obtain information that would otherwise require multiple cameras positioned at different corners. This copying approach maintains measurement precision while reducing device complexity and cost.
Solution Approach 2:
The reflector acts as an intermediary element that redirects light from the fingers to the single camera. By introducing this intermediary optical component, the system achieves multi-angle viewing capability without requiring multiple cameras, thus resolving the contradiction between measurement precision and device complexity.
2Device complexity
If a single camera is used to capture image windows, then the system cost is reduced, but the ability to accurately determine multiple touch points deteriorates due to occlusion
Solution Approach 1:
The patent introduces a temporal dimension by capturing successive image windows at different time points. By analyzing the movement of object images across multiple frames, the system can distinguish between occluded fingers and visible fingers, enabling accurate determination of multiple touch points even with a single camera. This transforms a spatial limitation into a solvable temporal problem.
Solution Approach 2:
The system performs preliminary capture of successive image windows before final gesture recognition. By collecting multiple frames in advance and analyzing the movement trajectories of object images, the system prepares the necessary data to resolve occlusion issues during gesture recognition, maintaining measurement precision while using a cost-effective single camera setup.
3Measurement precision
If the system calculates exact touch point coordinates based on object image positions, then precise positioning is achieved, but the system fails when pointers occlude each other
Solution Approach 1:
The patent transitions from static coordinate calculation to dynamic gesture recognition. Instead of relying on precise coordinate determination from single frames, the system analyzes the dynamic movement of object images across successive frames. This dynamic approach allows the system to recognize gestures based on movement patterns rather than absolute positions, maintaining reliability even when pointers occlude each other temporarily.
Solution Approach 2:
The system uses feedback from successive image windows to correct and verify gesture recognition. By continuously monitoring the movement of object images and comparing against expected gesture patterns, the system can distinguish between occlusion artifacts and actual gesture movements, thereby maintaining both measurement precision and reliability in the presence of occlusion.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate gesture recognition even when pointers occlude each other, reducing system costs by eliminating the need for multiple cameras and allowing for intuitive interactions like scrolling, scaling, and rotating without precise touch point positioning.
Implementation Method 1
an image sensor configured to capture an image window comprising object images of at least one pointer caused by that the pointer occludes the light emitting from the light source and/or the reflector
Implementation Method 2
an image sensor configured to capture an image window comprising object images of at least one pointer caused by that the pointer occludes the light emitting from the light source and/or the reflector
Data Source
AI summary
A gesture recognition method for an interactive system includes the steps of: capturing image windows with an image sensor; obtaining information of object images associated with at least one pointer in the image windows; calculating a position coordinate of the pointer relative to the interactive system according to the position of the object images in the image windows when a single pointer is identified according to the information of object images; and performing gesture recognition according to a relation between the object images in the image window when a plurality of pointers are identified according to the information of object images. The present invention further provides an interactive system.