Essential Matrix Generation Using Derived Point Pairs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating an essential matrix require detecting multiple feature point pairs between images, which can be cumbersome and inefficient, especially when dealing with fewer corresponding points.
Innovation Solution
An apparatus and method that detects three or more feature point pairs and derived point pairs, using feature values to determine distances and directions for deriving points, allowing for the generation of an essential matrix with fewer required feature point pairs, specifically using scale-invariant or affine-invariant feature values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple feature point pairs are detected between images using conventional methods, then the essential matrix can be generated with sufficient data, but the process becomes cumbersome and inefficient
Solution Approach 1:
The patent changes the parameter of feature point pair quantity from the conventional requirement of multiple pairs to only three pairs by introducing derived point pairs. This parameter change maintains the reliability of essential matrix generation while dramatically improving detection efficiency by reducing the number of feature points that need to be detected and matched between images.
Solution Approach 2:
The patent segments the feature point information by creating derived point pairs from existing feature point pairs. Instead of detecting all necessary point pairs directly, the method takes three detected feature point pairs and generates additional derived point pairs by applying geometric transformations, thereby obtaining sufficient data for essential matrix generation without detecting more original feature points.
2Productivity
If fewer feature point pairs are used for essential matrix generation, then the detection process becomes more efficient, but the accuracy and reliability may be compromised
Solution Approach 1:
The patent extends the problem from two dimensions (original feature points) to three dimensions by introducing derived point pairs that are geometrically transformed from the original pairs. This dimensional expansion allows the system to use only three original feature point pairs while generating additional constraint equations through their derived counterparts, maintaining accuracy while improving efficiency.
Solution Approach 2:
The patent performs preliminary geometric transformations to create derived point pairs from the detected feature point pairs before using them for essential matrix generation. By pre-computing these derived points with known spatial relationships, the method ensures that sufficient constraint information is available even when starting with fewer original feature point pairs, thus maintaining accuracy.
Data Source
AI summary
An essential matrix generation apparatus performs: detecting three or more feature point pairs from a first image and a second image; detecting, for each of two or more of the feature point pairs, a derived point pair that is a pair of a derived point separated by a first distance in a first direction from a point on the first image included in the feature point pair and a derived point separated by a second distance in a second direction from a point on the second image included in the feature point pair; generating an essential matrix representing an epipolar constraint on a point on the first image and a point on the second image by using the detected feature point pairs and derived point pairs.