Multimedia Complexity Analysis for On-Demand Broadcasting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for analyzing the complexity of multimedia content for on-demand broadcasting are resource-intensive and not suitable for live broadcasting, requiring multiple encodings to evaluate different formats, which leads to overload issues in OTT service providers and varying quality of service due to unmanaged networks.

Innovation Solution

A method that generates sets of multimedia content versions with specific resolutions, calculates video bitrate-quality values using predetermined functions based on temporal and spatial characteristics, and selects encoding profiles without actual encoding, allowing for targeted quality determination and profile selection suitable for live broadcasting.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple encodings are performed to evaluate different formats, then the quality of video streams is improved, but the resource consumption and system complexity increase significantly

Engineering Contradiction:
Improvequality of video streamsVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by performing complexity analysis on the source multimedia content before actual encoding. The system calculates temporal and spatial characteristics, determines overall complexity metrics, and pre-selects suitable encoding profiles based on this analysis. This avoids the need to perform multiple trial encodings to evaluate different formats, as the complexity analysis beforehand identifies which profiles will work best for the given content, thereby reducing both resource consumption and system complexity while maintaining reliable video quality.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies self-service by automatically analyzing the complexity of the source content and autonomously selecting appropriate encoding profiles without requiring manual intervention or multiple test encodings. The complexity analysis module self-evaluates the multimedia content's temporal and spatial characteristics and determines the optimal encoding configuration, eliminating the need for external quality assessment through multiple encoding attempts.

Inventive Principle:
Principle #25Self-service

2Reliability

If multiple encodings are performed to evaluate different formats, then the quality of video streams is improved, but the processing time and productivity are reduced

Engineering Contradiction:
Improvequality of video streamsVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies preliminary action by performing complexity analysis on the source multimedia content before actual encoding. The system calculates temporal and spatial characteristics, determines overall complexity metrics, and pre-selects suitable encoding profiles based on this analysis. This avoids the need to perform multiple trial encodings to evaluate different formats, as the complexity analysis beforehand identifies which profiles will work best for the given content, thereby reducing both resource consumption and system complexity while maintaining reliable video quality.

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If multiple encoding profiles are maintained for different network conditions, then the adaptability to network variations is improved, but the quantity of data to be managed increases

Engineering Contradiction:
Improveadaptability to network conditionsVSAvoidquantity of data to be managed
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent applies parameter changes by dynamically adjusting encoding parameters based on the complexity analysis results and current network conditions. Instead of maintaining separate encoding profiles for every possible scenario, the system calculates the overall complexity of the source content and uses this metric to select from a reduced set of profiles. The complexity analysis provides a continuous parameter that can map to different encoding configurations, allowing the system to adapt to varying network conditions while managing a limited set of encoding profiles.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system applies universality by creating a universal complexity analysis framework that works across different multimedia content types and network conditions. The temporal and spatial characteristic calculations provide a unified approach to assessing content complexity, which then informs profile selection for various scenarios. This single universal analysis method replaces the need for multiple specialized encoding profiles, reducing the quantity of data to be managed while maintaining adaptability.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Productivity

If complexity analysis is performed for live broadcasting, then the suitability for real-time delivery is improved, but the computational resources required increase

Engineering Contradiction:
Improvesuitability for real-time deliveryVSAvoidcomputational resources
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent applies taking out by extracting only the essential temporal and spatial characteristics from the multimedia content for complexity analysis, rather than performing full encoding or comprehensive quality assessment. The system extracts key metrics such as temporal complexity from motion estimation and spatial complexity from frame analysis, then uses these extracted features to determine overall complexity and select encoding profiles. This extraction approach provides sufficient information for real-time decision-making while minimizing computational resource requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies partial action by performing a simplified complexity analysis that focuses only on the most critical temporal and spatial characteristics needed for profile selection, rather than conducting exhaustive quality assessments. The analysis computes essential metrics like motion complexity and frame detail at reduced computational levels, which are sufficient for determining appropriate encoding profiles in real-time, thereby achieving suitable live broadcasting capability without excessive computational resource consumption.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3582500B1Method, system and computer program for analysing the complexity of multimedia content for on-demand broadcast
Publication Date: 2022.02.09 ATEME
  • EP3582500B1 patent drawingFigure 1
  • EP3582500B1 patent drawingFigure 2
  • EP3582500B1 patent drawingFigure 3

AI summary

The invention relates to a method implemented by computer means for digitally encoding source multimedia content into one or more video streams for on-demand distribution to at least one receiving terminal, the method comprising the following steps: - generating (S1) a set of versions (Vi) of said multimedia content, each version being characterized by a resolution and comprising a sequence of images, - determining (S2, S3), for each image in the sequence of images of each version, at least one temporal characteristic and at least one spatial characteristic, - calculating (S4), for each image in the sequence of images of each version, a range of video bitrate values, each video bitrate value being calculated by applying a predetermined function (F) to at least one temporal characteristic of said image,at least one spatial characteristic of said image and a value from a given range of theoretical values ​​relating to a target encoding quality so as to obtain, for each image, a set of pairs of theoretical video bitrate-quality values, - calculate (S5), for each version and for each value in the given range of theoretical encoding quality values, an overall video bitrate as a function of the video bitrates associated with said theoretical encoding quality value for each image in the sequence of images of said version so as to obtain, for each version, a set of pairs of overall video bitrate-theoretical quality values.