Parameterized Noise Model for AR Visual Coherence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality systems fail to adequately account for image noise when combining real and virtual content, leading to virtual content appearing detached or not fitting with real content due to lack of noise in virtual objects.
Innovation Solution
A method involving a processor to create a parameterized noise model based on noise data from captured images, generating a noise pattern to add noise to virtual content that matches the noise in real content, ensuring consistency and randomness similar to real noise patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If virtual content is rendered without noise to maintain clarity and computational efficiency, then rendering speed and processing efficiency are improved, but the virtual content appears detached and does not blend well with noisy real content
Solution Approach 1:
The patent applies parameter changes by introducing noise parameters to the virtual content rendering process. Instead of rendering clean virtual content, the system modifies the visual parameters of virtual content by adding synthesized noise that matches the statistical properties (variance, correlation) of noise in real captured content, enabling visual blending while maintaining rendering efficiency through parameterized noise models rather than full re-rendering
Solution Approach 2:
The patent uses an intermediary approach by introducing a noise synthesis component that acts as a mediator between the virtual content renderer and the final composite output. This intermediary generates noise patterns based on captured real content statistics and applies them to virtual content, bridging the visual gap without requiring changes to the core rendering pipeline or the real content capture process
2Stability of the object's composition
If noise is added to virtual content to match real content appearance, then visual coherence and realism are improved, but computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-computing noise statistics from captured real content during idle periods or initial setup, storing these statistical parameters for later use. This preliminary analysis of noise characteristics (variance, spatial correlation) allows the system to quickly generate matching noise for virtual content without performing complex real-time analysis during the actual rendering and compositing process
Solution Approach 2:
The patent uses copying by creating simplified statistical copies of the noise characteristics from real content rather than copying the actual noise pixels themselves. By extracting and storing noise parameters (mean, variance, correlation coefficients) and using these to generate synthetic noise for virtual content, the system replicates the essential visual properties of real noise without the computational burden of processing or storing large amounts of actual noise data
Data Source
AI summary
Various implementations disclosed herein render virtual content with noise that is similar to or that otherwise better matches the noise found in the images with which the virtual content is combined. Some implementations involve identifying noise data for an image, creating a parameterized noise model based on the noise data, generating a noise pattern approximating noise of the image or another image using the parameterized noise model, and rendering content that includes the image and virtual content with noise added based on the noise pattern.