Adaptive Bitrate Ladder Selection for Video Encoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video encoding techniques often fail to optimize bitrate and resolution effectively, leading to unacceptably poor visual quality or excessive resource usage when source data complexity differs from the 'typical' data used in designing the bitrate ladder.

Innovation Solution

A computer-implemented method for selecting a bitrate ladder by estimating the complexity of source data, assigning it to a corresponding complexity bucket, and choosing a bitrate ladder tailored to that complexity, optimizing tradeoffs between encoding quality and resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a fixed bitrate ladder is used for encoding, then the encoding process is simple and fast, but the visual quality deteriorates when source data complexity differs from typical data

Engineering Contradiction:
Improveencoding speedVSAvoidvisual quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies dynamics by transitioning from a static fixed bitrate ladder to a dynamic adaptive bitrate ladder selection system. The system evaluates source data complexity metrics (such as motion complexity, spatial complexity, and temporal complexity) and dynamically selects the most appropriate bitrate ladder from multiple available ladders. This allows the encoding process to adapt its parameters based on the actual characteristics of the input video, thereby maintaining high visual quality across diverse content types while preserving encoding efficiency.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements parameter changes by modifying the bitrate ladder selection based on measured source data complexity parameters. Instead of using a single fixed bitrate ladder, the system adjusts which bitrate ladder is selected by changing key parameters such as quantization parameter ranges, resolution settings, and bitrate allocations based on the evaluated complexity metrics. This parameter adaptation enables optimal visual quality for different content types without sacrificing encoding speed.

Inventive Principle:
Principle #35Parameter changes

2Productivity

If the bitrate ladder is tuned for simple cartoons, then encoding is efficient, but visual quality deteriorates for detailed action movies

Engineering Contradiction:
Improveencoding efficiencyVSAvoidvisual quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies local quality by creating different bitrate ladders tailored to specific content characteristics rather than using a universal ladder. Simple cartoons are encoded with ladders optimized for their low complexity (coarser quantization, lower bitrates), while detailed action movies receive ladders with finer quantization and higher bitrates appropriate for their complexity. This localized optimization ensures each content type receives the quality appropriate to its needs without wasting resources on overly complex encoding for simple content.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system dynamically selects between different pre-configured bitrate ladders based on real-time analysis of source data complexity. When simple content like cartoons is detected, the system selects the cartoon-optimized ladder for efficient encoding. When complex content like action movies is detected, it switches to the movie-optimized ladder. This dynamic selection resolves the contradiction by matching encoding parameters to content requirements.

Inventive Principle:
Principle #15Dynamics

3Manufacturing precision

If the bitrate ladder is tuned for detailed action movies, then visual quality is maintained, but resource consumption increases for simple cartoons

Engineering Contradiction:
Improvevisual qualityVSAvoidresource consumption
Core Design Contradiction:
Manufacturing precisionVSQuantity of substance

Solution Approach 1:

The patent changes encoding parameters (bitrate, resolution, quantization settings) based on source data complexity measurements. For simple cartoons, the system selects a bitrate ladder with lower bitrates and coarser quantization parameters, reducing resource consumption while maintaining adequate quality. For detailed action movies, it selects ladders with higher bitrates and finer quantization to preserve visual quality. This parameter adaptation eliminates the waste of using movie-optimized settings for simple content.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

Different bitrate ladders are created with locally optimized parameters for specific content types. The cartoon-optimized ladder uses parameters suited for simple content (lower bitrates, coarser quantization), while the movie-optimized ladder uses parameters for complex content (higher bitrates, finer quantization). By selecting the appropriate local optimization based on content analysis, the system avoids the resource waste that would occur from applying movie-optimized parameters to simple cartoons.

Inventive Principle:
Principle #3Local quality

4Quantity of substance

If lossy data compression is applied to increase compression rates, then storage and bandwidth are reduced, but visual quality deteriorates

Engineering Contradiction:
Improvedata sizeVSAvoidvisual quality
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent applies dynamics by making compression parameters adaptive rather than fixed. The system dynamically adjusts the degree of lossy compression (quantization parameter, bitrate allocation) based on source data complexity measurements. For simple content, more aggressive compression is applied with smaller data sizes and acceptable quality. For complex content, less aggressive compression is applied to maintain visual quality. This dynamic adaptation resolves the contradiction between compression ratio and visual quality.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes compression parameters (quantization parameter values, bitrate allocations, resolution settings) based on measured source data complexity. For simple cartoons, higher quantization parameters and lower bitrates are selected to achieve better compression with acceptable quality. For detailed action movies, lower quantization parameters and higher bitrates are selected to maintain visual quality. This parameter adaptation allows optimal balance between data size and visual quality for each content type.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3278561B1Techniques for optimizing bitrates and resolutions during encoding
Publication Date: 2020.07.22 NETFLIX INC
  • EP3278561B1 patent drawingFigure 1
  • EP3278561B1 patent drawingFigure 2
  • EP3278561B1 patent drawingFigure 3

AI summary

In one embodiment of the present invention, an encoding bitrate ladder selector tailors bitrate ladders to the complexity of source data. Upon receiving source data, a complexity analyzer configures an encoder to repeatedly encode the source data-setting a constant quantization parameter to a different value for each encode. The complexity analyzer processes the encoding results to determine an equation that relates a visual quality metric to an encoding bitrate. The bucketing unit solves this equation to estimate a bucketing bitrate at a predetermined value of the visual quality metric. Based on the bucketing bitrate, the bucketing unit assigns the source data to a complexity bucket having an associated, predetermined bitrate ladder. Advantageously, sagaciously selecting the bitrate ladder enables encoding that optimally reflects tradeoffs between quality and resources (e.g., storage and bandwidth) across a variety of source data types instead of a single, "typical" source data type.