Media Relay MCU Compressing Video Streams Without Decoding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current multipoint video conferencing systems require significant computational resources and are expensive due to the need for heavy processing to manage and decode multiple video streams, limiting the scalability and affordability of multipoint control units (MCUs) for conducting multiple conferencing sessions with composed continuous presence (CP) video images.

Innovation Solution

A Media Relay MCU (MRM) that receives compressed media from endpoints, determines which audio and video streams to relay without decoding or encoding, improving quality by reducing latency and processing demands, and allocates unique IDs to endpoints for efficient media management and distribution based on session properties and participant activity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a common MCU decodes and processes multiple video streams to manage continuous presence conferences, then the quality of composed video images is improved, but the computational resources and cost increase significantly

Engineering Contradiction:
Improvequality of composed video imagesVSAvoidcomputational resources
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the video processing function into two parts: the MCU performs only control and distribution of compressed streams, while the endpoints perform the actual video decoding and composition. This segmentation allows the MCU to avoid heavy computational processing while still enabling high-quality continuous presence conferences through coordinated endpoint processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary approach where the MCU acts as a control mediator rather than a processing intermediary. The MCU manages the conference control channel and distributes compressed video streams to endpoints, which then compose the final video images locally. This intermediary role reduction significantly decreases the computational burden on the MCU.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If a common MCU decodes and re-encodes audio and video streams for each participant, then compatibility with different endpoint properties is improved, but the processing time and computational power increase

Engineering Contradiction:
Improvecompatibility with different endpoint propertiesVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by having endpoints pre-decode video streams to their required resolution before composition. The MCU sends control information indicating which video streams to decode and at what resolution, allowing endpoints to prepare processed video frames in advance for efficient composition without real-time re-encoding delays.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent inverts the traditional processing model: instead of the MCU decoding and re-encoding streams for each participant, the endpoints themselves perform the decoding and composition of video streams. This inversion transfers the processing burden from the MCU to the endpoints, significantly reducing processing time and computational power requirements at the MCU.

Inventive Principle:
Principle #13The other way round (Inversion)

3Adaptability or versatility

If an MCU manages multiple compressed video streams with different resolutions, then the adaptability to various display requirements is improved, but the device complexity and cost increase

Engineering Contradiction:
Improveadaptability to various display requirementsVSAvoiddevice complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent applies local quality by allowing each endpoint to decode and compose video streams at its specific required resolution. Each endpoint processes video frames locally according to its display capabilities, with the MCU providing control information about which streams to process and at what quality levels. This eliminates the need for the MCU to maintain multiple resolution versions of each stream.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent implements self-service by enabling endpoints to autonomously decode, process, and compose video streams according to their own display requirements. The MCU provides minimal control guidance, but the endpoints independently handle the complex processing of multiple video streams with different resolutions, making each endpoint self-sufficient in adapting to its display capabilities.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2214410B1Method and system for conducting continuous presence conferences
Publication Date: 2019.10.09 POLYCOM INC
  • EP2214410B1 patent drawingFigure 1
  • EP2214410B1 patent drawingFigure 2
  • EP2214410B1 patent drawingFigure 3~4

AI summary

Disclosed are methods and systems for multipoint videoconferencing. A Media Relay MCU (MRM) (120) receives compressed media (audio, video, and/or data) from a plurality of endpoints (130) participating in a video conferencing session. For a given endpoint (130), the MRM (120) selects which of other endpoints (130) to display in a CP layout at the given endpoint (130). The MRM (120) transmits the compressed media from the selected endpoints to the given endpoint to be presented in the CP layout. The MRM (120) does not decode the compressed media.