Adaptive Feature Sharpening for Video See-Through XR Image Quality
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video see-through (VST) extended reality (XR) systems face challenges such as image noise and blur, which degrade the quality of the final views and user experience, due to limitations in image processing pipelines.
Innovation Solution
An adaptive feature sharpening technique is employed to enhance images by determining high-frequency features and applying a weighting map based on pixel blurriness, combining these features with the original pixels to generate enhanced images, which can be used in VST XR devices to improve image quality and reduce computational load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional image processing pipelines are used in VST XR systems, then device complexity is reduced, but image quality deteriorates due to noise and blur
Solution Approach 1:
The image processing pipeline is segmented into distinct modules: high-frequency feature extraction, blurriness detection, weighting map generation, and selective sharpening application. This modular approach improves image quality by addressing specific degradation factors separately while keeping each module computationally efficient.
Solution Approach 2:
The patent applies different processing intensities to different regions of the image based on local blurriness characteristics. The weighting map enables selective sharpening where needed while preserving natural appearance in already-clear regions, achieving local optimization of image quality without uniformly increasing processing complexity.
2Manufacturing precision
If uniform sharpening is applied to all pixels, then image quality improves, but noise is also amplified
Solution Approach 1:
The patent computes a weighting map that assigns different sharpening intensities to different pixels based on their local blurriness characteristics. Pixels with high blurriness receive stronger sharpening while pixels with low blurriness receive minimal or no sharpening, preventing noise amplification in already-clear regions.
Solution Approach 2:
The patent dynamically adjusts the sharpening parameter (weighting factor) for each pixel based on local image characteristics. The weighting map stores per-pixel or per-region weights that modulate the sharpening strength, allowing the system to adapt to local conditions and avoid uniform noise amplification.
3Manufacturing precision
If full-resolution image processing is performed, then image quality is maintained, but computational load increases
Solution Approach 1:
The patent applies sharpening processing selectively to only those regions of the image that require it, rather than processing the entire image uniformly. The weighting map identifies and targets specific areas with blurriness issues, reducing overall computational load while maintaining quality where needed.
Solution Approach 2:
The patent uses the weighting map to modulate processing intensity across different image regions. By varying the sharpening parameter based on local blurriness measurements, the system concentrates computational resources on problematic areas while using minimal processing in already-clear regions.
4Manufacturing precision
If aggressive sharpening is applied, then high-frequency features are enhanced, but artifacts are introduced
Solution Approach 1:
The patent applies different sharpening intensities to different regions based on local blurriness characteristics. The weighting map ensures that aggressive sharpening is applied only where necessary while using gentler processing elsewhere, preventing artifact introduction in regions that don't require strong enhancement.
Solution Approach 2:
The patent uses blurriness detection as feedback to control the sharpening process. The weighting map is generated based on measured blurriness levels, creating a closed-loop system that adjusts sharpening intensity according to actual image conditions, preventing over-sharpening and artifact generation.
Data Source
AI summary
An electronic device includes at least one processing device configured to obtain an image. The at least one processing device is also configured to determine high-frequency features of the image using the image. The at least one processing device is further configured to determine a weighting map based on blurriness of at least some pixels in the image, where the weighting map represents how much to sharpen the at least some pixels in the image. The at least one processing device is also configured to apply the weighting map to the high-frequency features of the image to generate weighted high-frequency features. In addition, the at least one processing device is configured to combine the weighted high-frequency features with the at least some pixels in the image to generate an enhanced image.