Mixed Reality Object Detection with Dynamic Color Model Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mixed reality systems face challenges in accurately and efficiently detecting real objects in varying lighting conditions due to color variations among users and the computational burden of real-time processing using neural networks, leading to instability and prolonged processing times.
Innovation Solution
An information processing apparatus that employs a two-tiered detection system: a first rapid detection using color information and a second, more accurate detection using trained models, with asynchronous updating of color information to stabilize object detection, reducing computational load and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If color-based detection is used for real-time object detection, then processing speed is improved, but detection stability deteriorates when object color changes due to optical conditions
Solution Approach 1:
The system dynamically switches between two detection modes: a fast color-based detection mode for real-time processing and a slower model-based detection mode for accuracy. The switching is triggered when color detection fails to meet reliability thresholds, allowing the system to adapt processing speed and stability requirements in real-time
Solution Approach 2:
A trained model serves as an intermediary verification layer that corrects and refines results from color-based detection. When color detection identifies objects or regions, the trained model validates and adjusts these results, ensuring detection stability even when color information is unreliable due to lighting changes
2Measurement precision
If trained models are used for accurate object detection, then detection precision is improved, but processing time increases
Solution Approach 1:
The system applies model-based detection selectively rather than continuously - only when color-based detection fails or when verification is needed. This partial application of the more accurate but slower method maintains precision while minimizing time loss by avoiding unnecessary model processing for clearly detectable objects
3Measurement precision
If neural network processing is performed for each frame, then detection accuracy is improved, but computational load and power consumption increase
Solution Approach 1:
The system uses periodic model-based verification at specific intervals or trigger conditions rather than continuous processing for every frame. Color-based detection runs continuously for efficiency, while model-based detection is periodically applied to verify results or correct errors, reducing overall computational load and power consumption while maintaining accuracy
Data Source
AI summary
An information processing apparatus includes at least one memory and at least one processor which function as: a first detection unit configured to repeatedly perform first processing for detecting a specific object from a real space by using setting information stored in a memory; a second detection unit configured to perform second processing for detecting the specific object from the real space, the second processing needing time longer than time needed for the first processing; an update unit configured to update the setting information, based on a result of the second processing; and a control unit configured to control to perform predetermined processing, based on a result of the first processing.