3D Motion Estimation Bias Correction in Multi-Camera Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating motion in multiple camera systems in 3D space are susceptible to bias, particularly in loop-less trajectories, and rely on auxiliary sensors or global optimization techniques.
Innovation Solution
A method that corrects 3D position estimations and distribution parameters using error propagation, calculates a bias direction, and adjusts motion parameters to reduce bias without relying on auxiliary sensors or loop-closing, thereby improving motion estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional motion estimation methods are used in multiple camera systems, then computation time is reduced, but bias in motion estimation increases particularly in loop-less trajectories
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing distribution parameters (mean and covariance matrices) for 3D positions of image features before motion estimation. These pre-computed statistical characteristics are then used during motion estimation to correct bias without requiring complex real-time calculations, thus improving accuracy while maintaining computation efficiency.
Solution Approach 2:
The patent changes parameters by transforming the motion estimation problem from minimizing 3D error to minimizing image space error, and by introducing corrected distribution parameters that account for stereo-reconstruction uncertainty. This parameter transformation allows bias correction in loop-less trajectories while maintaining computational efficiency through analytical solutions rather than iterative optimization.
2Measurement precision
If auxiliary sensors or global optimization techniques are used to reduce bias, then motion estimation accuracy improves, but device complexity increases
Solution Approach 1:
The patent applies self-service by enabling the multiple camera system to correct its own motion estimation bias using internally generated distribution parameters from stereo-reconstruction. The system uses its own 3D position data and uncertainty information to compute corrected motion parameters without external auxiliary sensors, thereby reducing system complexity while maintaining accuracy.
Solution Approach 2:
The patent introduces distribution parameters (mean and covariance matrices of 3D positions) as intermediaries between image features and motion estimation. These intermediary parameters capture stereo-reconstruction uncertainty and enable bias correction through analytical transformations, avoiding the need for complex global optimization or auxiliary sensors while improving measurement precision.
3Measurement precision
If loop-closing techniques are used to minimize drift, then trajectory accuracy improves, but ease of operation decreases due to additional processing requirements
Solution Approach 1:
The patent extracts the bias correction functionality from complex loop-closing procedures by isolating the essential statistical characteristics (distribution parameters) of 3D positions. This extraction allows bias correction to be performed independently through analytical means, eliminating the need for complex loop-closing operations while maintaining trajectory accuracy and simplifying processing.
Data Source
Figure 1
Figure 2a
Figure 2b
AI summary
The invention relates to a method of correcting a bias in a motion estimation of a multiple camera system in a three-dimensional (3D) space, wherein the fields of view of multiple cameras at least partially coincide. The method comprises the step of computing a first and second set of distribution parameters associated with corresponding determined 3D positions of image features in subsequent image sets. Further, the method comprises the step of estimating a set of motion parameters representing a motion of the multiple camera system. The method also comprises the steps of improving the computed first or second set of distribution parameters and improving the estimated set of motion parameters. Further, the method comprises calculating a bias direction based on the initially estimated set of motion parameters and on the improved estimated set of motion parameters.