Augmented Reality Global 3D Model Stitching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current augmented reality devices face limitations in capturing and rendering 3D models of large environments due to the limited sensor range and field of view of individual devices, leading to incomplete 3D mesh data and restricted AR functionality.
Innovation Solution
The system generates a global 3D model by combining local 3D models from multiple augmented reality devices, using external data to extend the depth perception and field of view, allowing for more comprehensive 3D interactions and immersive experiences across larger spaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If local 3D models are generated using individual device sensors, then the 3D model coverage is limited to the device's sensor range and field of view, but using multiple devices to expand coverage increases system complexity
Solution Approach 1:
The system divides the large environment into multiple local 3D models, each captured by individual devices within their sensor ranges. These segmented local models are then stitched together to form a comprehensive global 3D model, resolving the contradiction between limited individual device coverage and the need for large area coverage.
Solution Approach 2:
Multiple local 3D models from different devices are merged and stitched together to create a unified global 3D model. This combining approach enables extensive coverage area while managing system complexity by processing models in manageable segments rather than requiring all devices to capture the entire environment simultaneously.
2Measurement precision
If individual devices capture 3D data within their limited field of view, then the depth perception is restricted, but expanding field of view requires multiple devices which increases data processing complexity
Solution Approach 1:
The system segments the environment into multiple overlapping local 3D models, each captured with accurate depth information within the device's field of view. The segmentation allows each device to maintain high measurement precision for its local area while the overall system achieves extended coverage through systematic combination of these precise local models.
Solution Approach 2:
The system employs intermediary processing steps including feature matching, coordinate transformation, and model stitching to combine multiple local 3D models. These intermediary processes manage the data processing complexity by breaking down the integration task into manageable steps, enabling accurate depth perception across the global model without requiring all devices to process data simultaneously.
3Adaptability or versatility
If a global 3D model is created from multiple local models, then the AR functionality extends to larger spaces, but the time and computational resources required for model generation and stitching increase
Solution Approach 1:
The system performs preliminary actions by capturing and processing local 3D models incrementally as devices move through the environment, rather than requiring complete environment scanning before AR functionality begins. This allows the global model to be built progressively, enabling AR applications to start with available data and expand as more local models are acquired and stitched.
Solution Approach 2:
The global 3D model is dynamically constructed and updated as new local models are captured and integrated. The system adapts the global model in real-time based on device movement and new data acquisition, allowing AR functionality to expand continuously across larger spaces without requiring complete model generation upfront, thus reducing overall time loss.
Data Source
AI summary
In a device including a processor and a memory in communication with the processor is described, the memory includes executable instructions that, when executed by the processor, cause the processor to control the device to perform functions of: generating, based on a plurality of local 3D models, a global 3D model representing a portion of a real-world environment; determining a location of a 3D virtual object in the global 3D model; and generating augmentation data for rendering the 3D virtual object to be seen at a location of the real-world environment corresponding to the location of the 3D virtual object in the global 3D model.