Mixed Reality Position Estimation Using Environmental Parameter Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional mixed reality techniques experience a decrease in position estimation accuracy of image capturing units when the environment of the real space differs significantly between the time of map creation and the time of mixed reality experience.
Innovation Solution
An information processing apparatus that acquires images of the real space, extracts features, estimates the position of the image capturing unit, generates a map based on these features, and records environmental parameters to determine if the map can be used, notifying the user of potential accuracy issues if environmental changes exceed a set reference value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a map is created in advance regarding an experience area and self-position estimation is performed with reference to the map, then the processing load is reduced, but the position estimation accuracy decreases when the environment differs greatly between map creation and mixed reality experience
Solution Approach 1:
The patent applies preliminary action by creating a map in advance of the mixed reality experience. The map generation unit creates a map of the experience area before the user enters, storing feature information that will be used for subsequent position estimation. This allows the system to perform lightweight matching during the actual experience rather than heavy processing.
Solution Approach 2:
The patent applies parameter changes by introducing environmental parameters (illumination, temperature, humidity) as additional dimensions for comparing the current environment with the map creation environment. By monitoring these parameters and detecting significant changes, the system can determine when to trust the pre-created map for position estimation and when to regenerate it, thus maintaining accuracy while managing processing load.
2Measurement precision
If self-position estimation is performed with high accuracy, then the position estimation accuracy is improved, but the processing load becomes high
Solution Approach 1:
The patent applies partial action by performing position estimation only when necessary - specifically when environmental parameters indicate the environment has not changed significantly. The system partially uses the pre-created map for position estimation when conditions are favorable, avoiding full re-processing while maintaining accuracy when needed.
Solution Approach 2:
The patent applies feedback by continuously monitoring environmental parameters and using this information to determine whether to trust the pre-created map for position estimation. The environmental parameter acquisition unit provides feedback about environmental stability, which controls whether the position estimation unit uses the existing map or triggers map regeneration, thus balancing accuracy and processing load dynamically.
3Adaptability or versatility
If the environment of the real space differs greatly between map creation and mixed reality experience, then the adaptability to different environments is improved, but the position estimation accuracy decreases
Solution Approach 1:
The patent applies the intermediary principle by introducing environmental parameters as a mediator between the map data and position estimation process. These parameters act as an intermediary indicator that helps the system understand whether environmental conditions have changed significantly, allowing it to adapt to different environments while maintaining position estimation accuracy by regenerating maps only when necessary.
Solution Approach 2:
The patent applies parameter changes by monitoring environmental parameters (illumination, temperature, humidity) to detect significant environmental changes. When these parameters indicate substantial environmental differences, the system responds by regenerating the map to match current conditions, thus maintaining position estimation accuracy while adapting to different environments.
Data Source
AI summary
An information processing apparatus includes an image acquisition unit that acquires an image obtained by capturing a real space using an image capturing unit, an estimation unit configured to, based on a feature extracted from the image, estimate a position of the image capturing unit, a generation unit configured to, based on the feature and the estimated position of the image capturing unit, generate a map of the feature in the real space, a parameter acquisition unit that acquires an environmental parameter indicating an environment of the real space, a recording unit that records a first environmental parameter acquired at a first time point when the map is generated, in association with the map, and a determination unit configured to, based on the first environmental parameter and a second environmental parameter acquired at a second time point after the first time point, determine whether the map can be used.