Keypoint Descriptor Filtering by Orientation Angle
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing pipelines face challenges in efficiently determining matching keypoints in images, as this process typically requires significant CPU resources and increases power consumption.
Innovation Solution
The proposed solution involves filtering keypoint descriptors based on their orientation angles by selecting a subset of descriptors with matching or adjacent orientation ranges, thereby reducing the computational load by determining distances only between selected descriptors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If keypoint descriptors are compared exhaustively to determine matching keypoints, then matching accuracy is improved, but computational complexity and power consumption increase significantly
Solution Approach 1:
The patent segments the keypoint descriptor comparison process into two stages: first comparing orientation angle headers to identify candidate matches, then performing full descriptor comparison only on these candidates. This segmentation reduces the number of full comparisons needed while maintaining matching accuracy.
Solution Approach 2:
The patent performs preliminary filtering by comparing orientation angle headers before conducting full descriptor comparisons. This preliminary action eliminates non-matching descriptors early in the process, reducing computational complexity while preserving the ability to identify accurate matches.
2Measurement precision
If keypoint descriptors are compared exhaustively to determine matching keypoints, then matching accuracy is improved, but power consumption increases
Solution Approach 1:
The patent segments the power-consuming comparison operations into a low-power header comparison stage and a higher-power full descriptor comparison stage, executing only the necessary full comparisons on filtered candidates to reduce overall power consumption.
Solution Approach 2:
The patent performs preliminary filtering using orientation angle headers that require minimal computational resources and power, eliminating non-matching descriptors before more power-intensive full descriptor comparisons are performed.
3Speed
If a hardware image processing pipeline is implemented, then processing speed is improved, but device complexity increases
Solution Approach 1:
The patent segments the hardware processing into specialized units for header comparison and descriptor comparison, allowing parallel processing while keeping each unit relatively simple and focused on a specific task.
Solution Approach 2:
The patent extracts the orientation angle header comparison function as a separate preliminary processing step from the full descriptor comparison, allowing the hardware pipeline to efficiently filter candidates before detailed matching.
Data Source
AI summary
Embodiments of the present disclosure relate to selecting a subset of keypoint descriptors of two images for match operation based on their orientation angles indicated in headers of the keypoint descriptors. The keypoint descriptors in the two images are matched by first comparing their headers and then performing vector distance determination. During the header comparison operation, a header of a descriptor of a first image is compared only with headers of keypoint descriptors of a second image in a discrete orientation angle range corresponding to an orientation angle indicated by the header of the first image descriptor or keypoint descriptors of the second image in adjacent discrete orientation angle ranges. After the headers of the keypoint descriptors satisfying one or more matching criteria are determined, distance determination operations are performed between the keypoint descriptors while the remaining keypoint descriptors are discarded without determining their distances.