Multi-Axis Camera Self-Location Estimation Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-location estimation in real space can be inaccurate when images of visually similar scenes are captured from different locations and attitudes, leading to erroneous estimation of a target object's location and attitude.
Innovation Solution
An information processing apparatus and method that utilizes a plurality of image capturing units with different optical axes to estimate the location and attitude of a target object in real space, where a first image is captured by a main image capturing unit and verified by a second image capturing unit with a different optical axis, improving estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If self-location estimation is performed by comparing feature amounts from a single image capturing unit, then the estimation process is simple and fast, but accuracy degrades when visually similar scenes are captured from different locations and attitudes
Solution Approach 1:
The system divides the image capturing function into multiple independent image capturing units with different optical axes. Each unit captures images from its own perspective, and the system segments the verification process by using different units for estimation and verification, thereby improving accuracy without requiring a single complex capturing system
Solution Approach 2:
The patent introduces a verification mechanism using a second image capturing unit as an intermediary to check the estimation results. This intermediary unit provides an independent verification perspective, acting as a mediator between the estimation process and the final result, ensuring accuracy when visual similarity causes estimation errors
2Reliability
If multiple image capturing units with different optical axes are used for verification, then estimation accuracy is improved, but the device complexity and processing load increase
Solution Approach 1:
The system dynamically selects which image capturing unit serves as the estimation unit and which serves as the verification unit based on operational needs. This dynamic role assignment allows the system to maintain reliability through verification while managing complexity by flexibly configuring unit roles rather than requiring fixed complex architectures
Solution Approach 2:
The patent changes the operational parameters of the image capturing units by assigning different optical axes and roles (estimation vs. verification) to different units. This parameter differentiation enables the system to achieve reliable estimation and verification functions while managing device complexity through configurable parameter settings rather than fixed complex hardware
Data Source
AI summary
Provided is an information processing apparatus including an estimating unit that estimates at least one of a location or an attitude of a predetermined chassis in real space on the basis of a first image captured by a first image capturing unit among a plurality of image capturing units held in the chassis. The information processing apparatus further includes a verifying unit that verifies a likelihood of the estimation result on the basis of a second image captured by a second image capturing unit having an optical axis different from an optical axis of the first image capturing unit among the plurality of image capturing units.