Selective Image Pyramid Computation for Motion Blur Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Motion blur in visual tracking systems for AR and VR devices degrades tracking performance and increases computational requirements, particularly in head-worn devices with rapid camera movements, as it affects feature detection and matching accuracy.
Innovation Solution
A method to selectively apply the image pyramid process based on estimated or predicted motion blur levels using inertial data from IMUs or VIO systems, determining whether to downscale images to reduce computational intensity and maintain tracking accuracy without analyzing image pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the image pyramid process is applied to all images to mitigate motion blur, then tracking accuracy is improved, but computational requirements and power consumption increase
Solution Approach 1:
The patent changes the parameter of image processing intensity by selectively applying the image pyramid process based on motion blur detection. Instead of uniformly processing all images, the system dynamically adjusts processing levels according to detected motion blur conditions, thereby reducing overall computational load and power consumption while maintaining tracking accuracy when needed
Solution Approach 2:
The patent applies the image pyramid process partially rather than to all images. By detecting motion blur levels and selectively applying downscaling only to images exhibiting motion blur, the system performs partial processing that suffices to maintain tracking accuracy without the excessive computational cost of processing every image at full resolution
2Measurement precision
If the image pyramid process is applied to all images to mitigate motion blur, then tracking accuracy is improved, but computational operations increase
Solution Approach 1:
The system dynamically changes the processing parameter by applying different levels of the image pyramid process based on detected motion blur. Images with no or low motion blur are processed minimally or not at all, while images with high motion blur receive full processing, optimizing computational efficiency while maintaining tracking accuracy
Solution Approach 2:
The patent implements partial processing by applying the computationally intensive image pyramid process only to a subset of images that exhibit motion blur. This selective approach maintains tracking accuracy for critical frames while significantly reducing overall computational operations compared to processing all images
3Measurement precision
If images are downscaled using pyramid computation to reduce motion blur impact, then feature detection accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies downsampling partially only to images detected to have motion blur. By using motion blur detection to identify which images require processing, the system avoids unnecessary processing time on clear images while still achieving feature detection accuracy when motion blur is present
Solution Approach 2:
The system performs preliminary motion blur detection before applying the computationally intensive pyramid computation. This preliminary action identifies which images require downsampling, allowing the system to prepare and process only those specific frames that need it, thereby reducing overall processing time
Data Source
AI summary
A method for mitigating motion blur in a visual tracking system is described. In one aspect, a method for selective motion blur mitigation in a visual tracking system includes accessing a first image generated by an optical sensor of the visual tracking system, identifying camera operating parameters of the optical sensor during the optical sensor generating the first image, determining a motion of the optical sensor during the optical sensor generating the first image, determining a motion blur level of the first image based on the camera operating parameters of the optical sensor and the motion of the optical sensor, and determining whether to downscale the first image using a pyramid computation algorithm based on the motion blur level.