Adaptive Streaming Bit Rate Selection via Scene Complexity Maps

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Adaptive streaming solutions face challenges when using variable bit rate (VBR) video streams, as the average bit rate does not reflect the bit rate of specific scenes, leading to inappropriate switching and reduced playback quality.

Innovation Solution

A method that involves receiving a scene complexity map to dynamically select encoded video streams based on scene complexity levels, allowing for optimal playback by switching between constant bit rate streams during playback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If variable bit rate encoding is used to optimize bandwidth utilization, then file size and bandwidth usage are reduced, but the average bit rate does not reflect the bit rate of specific scenes leading to inappropriate stream switching

Engineering Contradiction:
Improvebandwidth usageVSAvoidbit rate representation accuracy
Core Design Contradiction:
Loss of energyVSMeasurement precision

Solution Approach 1:

The video stream is divided into multiple segments or scenes, each with its own complexity assessment. The system segments the video content and evaluates each segment's complexity independently, allowing for more precise bit rate selection for each portion rather than using a single average bit rate for the entire stream.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different portions of the video stream are assigned different quality levels based on their complexity. High-complexity scenes receive higher bit rates while low-complexity scenes use lower bit rates, optimizing bandwidth usage while maintaining overall quality. This local quality adjustment resolves the contradiction by making bit rate allocation scene-specific rather than stream-wide.

Inventive Principle:
Principle #3Local quality

2Adaptability or versatility

If multiple encoded video streams are prepared for adaptive streaming, then playback quality can be adapted to network conditions, but the complexity of selecting the appropriate stream increases

Engineering Contradiction:
Improvenetwork condition adaptationVSAvoidstream selection complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The complexity assessment and segmentation of video content is performed in advance during encoding, before playback. This preliminary action creates a complexity map that guides stream selection during playback, reducing the real-time decision complexity while maintaining adaptability to network conditions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from network condition monitoring combined with pre-computed complexity information to dynamically select appropriate video streams. This feedback mechanism simplifies the selection process by providing clear guidance based on both content requirements and current network capabilities.

Inventive Principle:
Principle #23Feedback

3Manufacturing precision

If video streams are encoded to different bit rates for adaptive streaming, then playback quality can be optimized for different bandwidth conditions, but switching between streams may not be appropriate when average bit rate does not reflect scene complexity

Engineering Contradiction:
Improveplayback qualityVSAvoidstream switching appropriateness
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent applies local quality by assigning different quality levels (bit rates) to different portions of the video stream based on their complexity. This ensures that each scene receives appropriate quality treatment, making stream switching more appropriate and reliable.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system changes the bit rate parameter dynamically based on scene complexity assessments. By adjusting the bit rate parameter to match the actual complexity of each video portion rather than using average bit rates, the system improves both playback quality and the appropriateness of stream switching decisions.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2649599B1Variable bit video streams for adaptive streaming
Publication Date: 2018.06.13 NETFLIX INC
  • EP2649599B1 patent drawingFigure 1
  • EP2649599B1 patent drawingFigure 2
  • EP2649599B1 patent drawingFigure 3

AI summary

One embodiment of the present invention sets forth a technique for adapting playback bit rate in a content delivery system based on scene complexity of the video content as well as network conditions and other performance factors. A scene complexity map of the video content indicates the complexity levels of different scenes within the video content. Using the scene complexity map, a content player may download scenes of lower scene complexity levels from video streams encoded to lower bit rates to manage the bandwidth consumed to download the video content and allow for the downloading of scenes of higher complexity levels from video streams encoded to higher bit rates.