Spherical Video Denoising via Adaptive Noise Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Denoising spherical videos is challenging due to spatially varying noise characteristics caused by camera geometry and map projections, which are often unknown, leading to artifacts when encoding at low bitrates, increasing storage and network costs.
Innovation Solution
A method and apparatus for denoising spherical videos by estimating noise models for frame blocks, adjusting these models based on adjacent block characteristics, and using them to denoise the video content, with different techniques employed depending on the availability of camera geometry and map projections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If video content is encoded at low bitrates to reduce storage and network costs, then storage and network costs are reduced, but artifacts are introduced due to spatially varying noise characteristics
Solution Approach 1:
The patent applies denoising processing to video content before encoding. By removing spatially varying noise characteristics in advance, the video can be encoded at lower bitrates without introducing artifacts, thus resolving the contradiction between reducing storage costs and maintaining video quality.
Solution Approach 2:
The patent estimates separate noise models for different frame blocks based on their local characteristics. This local approach allows targeted denoising that preserves important video features while removing noise, enabling efficient compression without quality loss in specific regions.
2Manufacturing precision
If noise models are estimated for each frame block to improve denoising accuracy, then denoising effectiveness is improved, but processing complexity increases
Solution Approach 1:
The patent divides the video content into frame blocks and estimates noise models for each block independently. This segmentation allows accurate local denoising while enabling parallel processing, which manages computational complexity through distributed computation.
Solution Approach 2:
The patent focuses noise model estimation on representative frame blocks and uses these to infer characteristics of adjacent blocks. This partial approach reduces the total number of noise models that need to be explicitly estimated while maintaining denoising accuracy across the entire video.
Data Source
AI summary
Processing a spherical video using denoising is described. Video content comprising the spherical video is received. Whether a camera geometry or a map projection, or both, used to generate the spherical video is available is then determined. The spherical video is denoised using a first technique responsive to a determination that the camera geometry, the map projection, or both is available. Otherwise, the spherical video is denoised using a second technique. At least some steps of the second technique can be different from steps of the first technique. The denoised spherical video can be encoded for transmission or storage using less data than encoding the spherical video without denoising.