AR Toy System Sub-Image Processing for Power Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Toy systems for augmented reality face challenges in reducing processing power requirements for detecting user manipulations in dynamic real-world scenes, leading to increased power consumption and potential for undesired object detection, especially in cluttered environments, and require a solution that is user-friendly for children without programming knowledge.
Innovation Solution
A toy system that uses a data processing unit with an image capturing device and display to recognize reference toys, processing only sub-images depicting predetermined reference positions to detect user manipulations, thereby reducing processing resources and minimizing false detections, while prompting users to interact at specific points through computer-generated feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the AR system processes the entire captured image to detect user manipulations, then detection accuracy is improved, but processing power consumption increases
Solution Approach 1:
The patent divides the captured image into multiple sub-images corresponding to different reference positions where toys may be placed. The system processes only these specific sub-images rather than the entire image, reducing processing power consumption while maintaining detection accuracy for toy manipulations at reference positions.
2Reliability
If the AR system processes the entire captured image to detect user manipulations, then detection completeness is improved, but processing time increases
Solution Approach 1:
The patent segments the image processing task by identifying and processing only the sub-images at predetermined reference positions. This segmentation reduces processing time significantly while maintaining detection completeness for toys placed at these known reference positions, as the system doesn't need to analyze the entire image.
3Difficulty of detecting and measuring
If the AR system detects all objects in the real-world scene, then object detection capability is improved, but false detections increase in cluttered environments
Solution Approach 1:
The patent applies local quality by focusing detection efforts specifically on predetermined reference positions where toys are expected to be placed. By concentrating processing resources on these specific locations rather than analyzing all objects in the scene, the system reduces false detections in cluttered environments while maintaining effective toy detection capability.
4Productivity
If the system processes only sub-images at reference positions, then processing efficiency is improved, but detection coverage may be reduced
Solution Approach 1:
The patent uses preliminary action by pre-defining reference positions where toys are expected to be placed based on the AR game design. This allows the system to efficiently process only relevant sub-images at these predetermined locations, maintaining both high processing efficiency and adequate detection coverage for the intended toy interaction scenarios.
Data Source
AI summary
A toy system, the toy system comprising a data processing system and one or more toys, the plurality of toys including at least one reference toy having a visual appearance recognisable by the data processing system in a captured image of a real-world scene including said at least one reference toy, the data processing system comprising an image capturing device, a processing unit, and a display, wherein the data processing system has stored thereon information associated with the at least one reference toy, the information including at least one predetermined reference position defined relative to the at least one reference toy; and wherein the data processing is configured to: capture a sequence of images of a real-world scene, the real-world scene including said at least one reference toy; process the captured images to detect and recognize said at least one reference toy within the real-world scene; retrieve the at least one predetermined reference position from the stored information associated with the recognized at least on reference toy; process the captured images to identify the at least one predetermined reference position within at least a first image of the sequence of captured images; selectively process a sub-image within the first image, the sub-image depicting said identified reference position to selectively detect a user manipulation of the real-world scene at the identified at least one predetermined reference position; responsive to detecting the user manipulation, generate and render computer-generated perceptual information associated with the detected user manipulation.