Key Frame Distribution for Video Conferencing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In multi-party video conferencing, switching video information resolution can cause delays and network congestion due to the need to transmit key frames to all endpoints, especially when active status changes, leading to interruptions and suboptimal viewing experiences.

Innovation Solution

A server extracts key frames from non-active video streams during times of excess bandwidth and stores them for later transmission to endpoints, allowing for faster resolution switching and reducing network congestion by anticipating which streams will become active.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If key frames are transmitted to all endpoints when active status changes, then video resolution switching is enabled, but network congestion and delays occur

Engineering Contradiction:
Improvevideo resolution switching capabilityVSAvoiddelay in video playback
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The server performs preliminary action by extracting and buffering key frames from video streams of non-active participants before they are needed. When a participant transitions to active status, the pre-buffered key frames are immediately transmitted to endpoints, eliminating the delay that would otherwise occur while waiting for key frames to be generated and transmitted.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The video stream is segmented into key frames and non-key frames. The server selectively buffers only key frames for non-active participants while transmitting only non-key frames during their inactive period. This segmentation allows endpoints to quickly switch to a different participant's stream by retrieving pre-buffered key frames without receiving the entire video stream.

Inventive Principle:
Principle #1Segmentation

2Reliability

If key frames are transmitted to all endpoints, then video quality is maintained, but bandwidth consumption increases

Engineering Contradiction:
Improvevideo qualityVSAvoidbandwidth consumption
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

Different quality levels are provided locally to different participants based on their active status. Active participants receive full-quality video streams, while non-active participants have only their key frames buffered at the server. This local differentiation of quality ensures that bandwidth is consumed only when necessary for maintaining video quality at the receiving end.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The server performs partial action by buffering only key frames rather than complete video streams for non-active participants. This partial buffering provides sufficient information for quick stream switching without the excessive bandwidth consumption of buffering entire video streams, achieving a balance between video quality and bandwidth usage.

Inventive Principle:
Principle #16Partial or excessive action

3Stability of the object's composition

If complete video streams are transmitted for all participants, then video continuity is ensured, but network congestion occurs during active status changes

Engineering Contradiction:
Improvevideo stream continuityVSAvoidnetwork efficiency
Core Design Contradiction:
Stability of the object's compositionVSProductivity

Solution Approach 1:

The video transmission system dynamically adapts its behavior based on participant active status. For non-active participants, the server transmits only periodic key frames and buffers them, rather than continuously transmitting complete video streams. When a participant becomes active, the system dynamically switches to transmitting their complete video stream. This dynamic approach maintains video continuity when needed while improving network efficiency during inactive periods.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS8917309B1Key frame distribution in video conferencing
Publication Date: 2014.12.23 GOOGLE LLC
  • US8917309B1 patent drawing
  • US8917309B1 patent drawing
  • US8917309B1 patent drawing

AI summary

A system, apparatus, and method for relaying video information that is part of a multi-party video communication session having multiple endpoints. The server can receive multiple video information streams from multiple endpoints and re-transmit some or all of the video information streams to the endpoints with one or more of the video information steams identified as having active status and being decoded at high resolution at the endpoints, while transmitting key frames from video information streams not having active status. Upon switching active speaker status to a new video information stream, the endpoints, already having a key frame from the switched-to video information, can switch to decoding the new video information stream having active status at high resolution without delay.