SLAM Key Frame Extraction Using Adaptive Motion Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SLAM technologies face issues of key frame redundancy and uneven spatial distribution, leading to memory consumption and low positioning accuracy due to the use of fixed thresholds and insufficient feature points during key frame extraction.
Innovation Solution
A method and device for extracting key frames in SLAM that utilize adaptive thresholds calculated based on relative motion information between image frames, selecting key frames based on feature points and spatial information to avoid redundancy and ensure even distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed threshold is used for key frame extraction, then the extraction process is simple, but key frame redundancy occurs and memory consumption increases
Solution Approach 1:
The patent applies dynamics by transforming the fixed threshold into an adaptive threshold that dynamically adjusts based on the current image frame's characteristics. The threshold is no longer static but changes according to the relative motion information and feature point quality of each frame, allowing the system to adapt to varying scene conditions and avoid redundant key frame selection.
Solution Approach 2:
The patent changes the parameter of the threshold from a fixed value to a dynamically calculated value based on multiple factors including relative motion information, feature point quantity, and feature point quality. This parameter change enables the threshold to adapt to different scenarios, reducing key frame redundancy while maintaining extraction effectiveness.
2Measurement precision
If key frames are extracted frequently to improve positioning accuracy, then positioning accuracy improves, but key frame spatial distribution becomes uneven and memory consumption increases
Solution Approach 1:
The patent introduces multiple dynamic parameters including relative motion information, feature point quantity, and feature point quality to control key frame extraction. By adjusting these parameters adaptively, the system achieves better positioning accuracy while maintaining more uniform spatial distribution of key frames compared to frequent fixed-threshold extraction.
Solution Approach 2:
The patent implements feedback mechanisms by evaluating the current image frame's characteristics (relative motion, feature point quantity, feature point quality) and using this feedback to adjust the threshold for key frame extraction. This feedback loop ensures that key frames are selected based on actual scene changes rather than fixed intervals, improving both positioning accuracy and spatial distribution uniformity.
3Device complexity
If a fixed threshold is used for key frame extraction, then the extraction method is simple, but feature point quality is insufficient and positioning accuracy decreases
Solution Approach 1:
The patent transforms the simple fixed threshold into a composite adaptive threshold that incorporates multiple parameters: relative motion information, feature point quantity, and feature point quality. This parameter transformation enables the extraction method to identify higher quality feature points and select more accurate key frames, directly improving positioning accuracy.
Solution Approach 2:
The patent applies dynamics by making the threshold adaptive rather than static. The threshold dynamically responds to the quality characteristics of each image frame, allowing the system to selectively extract key frames with sufficient feature point quality, thereby improving positioning accuracy without excessive complexity.
4Quantity of substance
If adaptive threshold based on relative motion information is used, then key frame redundancy is reduced, but calculation complexity increases
Solution Approach 1:
The patent changes the threshold parameter from fixed to adaptive based on relative motion information. While this increases calculation complexity compared to a fixed threshold, it significantly reduces key frame quantity by avoiding redundant selections, achieving a favorable trade-off between computational effort and output efficiency.
Data Source
AI summary
A method and device for extracting key frames in simultaneous localization and mapping and a smart device. The method includes acquiring an image frame from an image library storing a plurality of image frames of an unknown environment, and performing feature extraction on the image frame to obtain information of feature points, wherein the information includes a quantity of feature points; acquiring relative motion information of the image frame relative to the previous key frame, and calculating an adaptive threshold currently used by using the relative motion information; and selecting a key frame according to the information of feature points and the adaptive threshold indicating space information of image frames.