Adaptive Loop Closure Detection for Camera Motion Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional loop closure detection (LCD) in smart eyewear and other devices with cameras is inefficient due to unnecessary running at fixed periodic rates, leading to false alarms, increased power consumption, and missed events.
Innovation Solution
Adaptive loop closure detection (LCD) is implemented, which is event-driven and based on user motion, user interactions, and specific AR application decisions to optimize when to enable or disable LCD.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If conventional loop closure detection methods are used, then the system can operate, but the computational complexity increases and processing speed decreases due to comparing current sensor data with all historical sensor data
Solution Approach 1:
The patent segments the large historical sensor dataset into multiple data blocks, each representing a specific time period or spatial region. Instead of comparing current sensor data with all historical data at once, the system divides the comparison task into smaller segments, significantly reducing computational complexity and improving processing speed while maintaining detection accuracy.
2Reliability
If sensor data is collected continuously for loop closure detection, then detection accuracy is maintained, but energy consumption increases
Solution Approach 1:
The patent implements periodic action by collecting and processing sensor data at specific intervals rather than continuously. The system determines when loop closure detection should be performed based on predefined criteria such as time intervals, distance thresholds, or location markers, thereby maintaining detection accuracy while significantly reducing energy consumption during operation.
3Measurement precision
If all sensor data is processed for loop closure detection, then detection precision is improved, but the time required for detection increases
Solution Approach 1:
The patent applies preliminary action by pre-processing and organizing sensor data into structured blocks during idle periods or when data is naturally available. Historical sensor data is pre-segmented, pre-filtered, and pre-organized into manageable blocks with associated metadata. When loop closure detection is needed, the system can quickly query and compare against pre-organized data blocks, maintaining high detection precision while minimizing detection time.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method of performing loop closure detection is described. The method comprises detecting a movement of a device having a camera; and adaptively disabling or enabling, using a processor of the device, a loop closure detection of the device based upon the detected movement of the device.