Bit Rate Ladder Creation Using Quality Reference Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video streaming technologies face challenges in adapting bit rates to available bandwidth without significant quality loss, leading to inefficient data transmission and storage.
Innovation Solution
A method and device for creating a bit rate ladder by determining reference points based on quality requirements, using a neural network to interpolate and process qualities, and selecting a subset of reference points to minimize quality differences while optimizing encoding and storage costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple versions of the same video with different quality levels are provided to adapt to different bandwidths, then the adaptability to network conditions is improved, but the device complexity and storage requirements increase due to maintaining multiple representations
Solution Approach 1:
The patent segments the continuous quality-bw space into discrete reference points and quality levels. By dividing the bit rate ladder into specific segments (reference points at different quality levels) rather than treating it as a continuous spectrum, the system achieves adaptability to different bandwidths while managing complexity through structured segmentation of the representation space
Solution Approach 2:
The patent changes the parameter of quality measurement from subjective or complex metrics to objective quality measures that can be systematically varied across reference points. By parameterizing quality at different bit rates and resolutions with objective metrics, the system enables adaptability while controlling complexity through parameter-based management of representations
2Manufacturing precision
If a large number of reference points are used to ensure high quality across all bandwidths, then the image quality is improved, but the encoding costs and processing time increase
Solution Approach 1:
The patent extracts only the essential reference points needed to define quality levels rather than using a dense grid of all possible bit rate and resolution combinations. By taking out and selecting only the critical reference points that sufficiently characterize quality across bandwidths, the system maintains image quality while reducing encoding costs and processing requirements
Solution Approach 2:
The patent uses a sufficient but not excessive number of reference points. Rather than exhaustively testing all possible encoding parameters, the method employs a partial set of reference points that provides adequate quality coverage. This partial action approach achieves the necessary image quality without the excessive encoding effort that would result from comprehensive parameter testing
3Adaptability or versatility
If the bit rate is reduced to accommodate limited bandwidth, then the network transmission capability is improved, but the image quality deteriorates
Solution Approach 1:
The patent creates a dynamic bit rate ladder where the selection of representations is adaptable to available bandwidth. By establishing a structured set of reference points across different quality levels, the system can dynamically select appropriate representations based on network conditions, ensuring bandwidth compatibility while maintaining the highest possible quality for each condition
4Measurement precision
If comprehensive quality testing is performed at all bit rate and resolution combinations, then the accuracy of quality assessment is improved, but the measurement time and computational resources increase
Solution Approach 1:
The patent performs preliminary quality assessment at a subset of reference points before finalizing the bit rate ladder. By conducting measurements in advance at strategically selected reference points and using these results to define quality levels, the system achieves accurate quality assessment without the time cost of exhaustive testing at all possible parameter combinations
Data Source
AI summary
The present disclosure relates to methods and devices for creating a bit rate ladder for encoding representations of a video section. Creating includes generating a set of reference points, where a reference point indicates a quality of a representation based on the bit rate and the resolution. A subset of reference points is selected to create the bit rate ladder while taking quality requirements into account.


