3D Input Space Pattern Recognition for Mixed Reality Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented interactive input systems are limited by the ability to only track the position and orientation of physical objects, making interactions like clicking buttons or opening menus difficult without external devices, and suffer from issues like temporary loss of tracking and marker identification problems under poor lighting conditions.
Innovation Solution
An interactive input system with a 3D input space that uses a computing structure and input devices to recognize patterns within a 3D space, allowing for the detection of electronic devices and modification of digital content based on their state, and includes features like pattern recognition, transition effects, and robust marker identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If physical markers are used for tracking, then pattern recognition is enabled, but marker identification fails under poor lighting conditions
Solution Approach 1:
The patent employs color-coded patterns on physical markers that can be reliably identified under various lighting conditions. The system uses distinct color combinations and patterns that maintain their distinguishability even when lighting varies, allowing the camera to accurately recognize marker states without being significantly affected by ambient light changes.
2Adaptability or versatility
If multiple patterns are required on a marker, then interaction capabilities are enhanced, but the marker becomes difficult to manipulate
Solution Approach 1:
The patent transitions from 2D marker patterns to 3D physical objects with multiple visible faces, each displaying a pattern. This dimensional change allows the system to encode multiple interaction states across different faces of the same object, enabling complex interactions while maintaining a compact, easily manipulable form factor.
Solution Approach 2:
The marker is divided into multiple faces, with each face containing a distinct pattern. This segmentation allows the system to recognize different states based on which face is currently visible to the camera, enabling multiple interaction modes without requiring a single large complex pattern that would be difficult to manipulate.
3Device complexity
If only position and orientation tracking is implemented, then system complexity is reduced, but interaction capabilities are limited
Solution Approach 1:
The patent enhances tracking by analyzing local pattern characteristics on different faces of the physical object, not just global position and orientation. By recognizing specific patterns on specific faces and their spatial relationships, the system derives additional interaction information (such as which face is active) without significantly increasing overall system complexity.
4Measurement precision
If pattern size is increased to improve recognition, then recognition reliability improves, but the marker becomes difficult to manipulate
Solution Approach 1:
The patent uses the third dimension (multiple faces of a 3D object) to provide additional pattern recognition opportunities. Each face can have a clearly visible pattern of appropriate size for reliable recognition, while the compact 3D form factor maintains ease of manipulation. The system determines state based on which face is visible, eliminating the need for oversized single-face patterns.
Data Source
AI summary
An interactive input system comprises computing structure; and an input device detecting at least one electronic device displaying a recognizable pattern within a three-dimensional (3D) input space and providing output to the computing structure. The computing structure processes the output of the input device to recognize the pattern displayed by the at least one electronic device in the 3D input space; detect the state of the at least one electronic device; and modify image data by applying a transition to digital content associated with the displayed pattern based on the detected state of the at least one electronic device.