User Pose Estimation Using 3D Virtual Space Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating user pose using spatial maps are sensitive to spatial information acquisition poses, leading to reduced precision due to system load issues, discontinuous spatial maps, and temporal changes in the real space, such as lighting and object movements, which can degrade the accuracy of user pose estimation.
Innovation Solution
A method and apparatus that utilize a three-dimensional virtual space model constructed from depth and image information to estimate user pose by distinguishing background and non-background areas, generating corresponding information, and calculating similarity between user and model information, allowing for robust pose estimation despite changes in the real space over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spatial information is acquired from multiple poses to reduce sensitivity, then measurement precision improves, but system load increases due to acquisition time, capacity, and processing speed requirements
Solution Approach 1:
The patent performs preliminary actions by acquiring spatial information from multiple poses in advance and constructing a comprehensive spatial map before user pose estimation is needed. This pre-processing approach stores multiple views of the environment, allowing the system to later select from pre-acquired data rather than acquiring new data during estimation, thus reducing real-time system load while maintaining precision
Solution Approach 2:
The patent applies partial action by selectively using only the necessary portion of pre-acquired spatial information from multiple poses. Instead of processing all possible pose data, the system chooses a subset sufficient for accurate estimation, balancing between acquiring enough diverse views to reduce sensitivity and limiting the amount of data to manage system load
2Productivity
If spatial information is acquired from a few poses to reduce system load, then processing efficiency improves, but the spatial map cannot fully express the real space, degrading measurement precision
Solution Approach 1:
The patent uses partial action optimally by acquiring spatial information from a limited but strategically selected number of poses. This selective approach captures sufficient environmental features to create an accurate spatial map while keeping the data volume manageable for efficient processing, achieving the right balance between completeness and efficiency
3Device complexity
If point cloud is used to construct spatial map, then device complexity is reduced, but discontinuous spatial map is generated, degrading measurement precision
Solution Approach 1:
The patent merges multiple point cloud datasets acquired from different poses into a unified spatial map. By combining these partial point clouds through registration and fusion algorithms, the system creates a more complete and continuous representation of the environment, reducing gaps and discontinuities while maintaining relative simplicity compared to other 3D reconstruction methods
4Reliability
If spatial map is updated frequently to reflect real space changes, then reliability improves, but system load and processing time increase
Solution Approach 1:
The patent implements periodic action by updating the spatial map at predetermined time intervals rather than continuously or on every change. This periodic update strategy maintains the spatial map's reliability by regularly refreshing it with current environmental data while avoiding the excessive processing load of continuous updates, achieving a balanced refresh rate
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The approach reduces sensitivity to spatial information acquisition poses, improves processing efficiency, and maintains precision even with changes in the real space, enhancing user immersion in mixed reality applications.
Implementation Method 1
a depth measurement device that measures depth information
Implementation Method 2
an image acquisition device that acquires image information
Data Source
AI summary
The present invention relates to a method and an apparatus for estimating a user pose using a three-dimensional virtual space model. The method of estimating a user pose including the position and orientation information of a user for a three-dimensional space includes a step of receiving user information including an image acquired in a three-dimensional space, a step of confirming a three-dimensional virtual space model constructed based on spatial information including depth information and image information for the three-dimensional space, a step of generating corresponding information corresponding to the user information in the three-dimensional virtual space model, a step of calculating similarity between the corresponding information and the user information, and a step of estimating a user pose based on the similarity.