Video Masking Windows for Scene-Cut Bitrate Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques face challenges in reducing bitrate for high-resolution video content without compromising perceived quality and maintaining reasonable computational complexity.
Innovation Solution
Implementing a joint backward and forward temporal masking technique that dynamically adjusts quantization values based on the proximity to scene cuts in video content, utilizing masking windows with higher quantization values near scene cuts and lower values further away, to minimize artifacts undetectable by the human visual system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If video compression techniques are applied to reduce bandwidth, then bitrate is reduced, but computational complexity increases
Solution Approach 1:
The video content is segmented into different portions based on temporal masking windows, where frames are divided into regions near scene cuts (first portion) and regions further away (second portion). Different quantization values are applied to different segments, allowing optimized compression without uniformly increasing computational complexity across the entire video stream.
Solution Approach 2:
Different quantization values are applied locally to different portions of the video content. Higher quantization values are used for frames in the first portion (near scene cuts) where artifacts are less noticeable, while lower quantization values are used for frames in the second portion where quality is more critical. This local differentiation reduces overall computational complexity while maintaining perceived quality.
2Productivity
If quantization values are increased to reduce bitrate, then compression efficiency improves, but visual quality deteriorates
Solution Approach 1:
The quantization values are dynamically adjusted based on the temporal position of frames relative to scene cuts. The system determines which frames fall into the first portion (near scene cuts) and which fall into the second portion, then applies different quantization values accordingly. This dynamic adaptation allows higher compression efficiency in regions where artifacts are less noticeable while maintaining visual quality where needed.
Solution Approach 2:
The quantization parameter is changed based on the temporal masking window analysis. By analyzing scene cuts and determining appropriate masking windows, the system changes the quantization parameter dynamically - using higher values for frames where they provide greater compression benefit and lower values for frames where visual quality is more important, thus optimizing the balance between compression efficiency and visual quality.
3Productivity
If frame removal is applied near scene cuts to reduce bitrate, then compression efficiency improves, but artifacts become more noticeable
Solution Approach 1:
The system exploits the fact that human visual systems are less sensitive to artifacts near scene cuts - a characteristic that could be considered a vulnerability in traditional compression approaches. By applying higher quantization values (and thus allowing more frame removal) specifically in the first portion near scene cuts, the patent converts this vulnerability into a benefit, achieving greater compression efficiency in regions where artifacts are least noticeable to viewers.
Data Source
AI summary
A first quantization value for encoding at least one frame of a content item may be determined based at least on a predetermined bitrate and a point in the content item associated with a scene change. A first duration associated with a first portion of the content item may be determined. The first portion of the content item may comprise the at least one frame and may be associated with the first quantization value. A second quantization value for encoding at least another frame of the content item may be determined based at least on the predetermined bitrate. A second duration associated with a second portion of the content item may be determined. The second portion of the content item may comprise the at least another frame and may be associated with the second quantization value.