Audio-Video Synchronization Using Motion Correlation Vectors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing audio-video synchronization techniques require significant computational resources and can be sub-optimal due to inaccuracies in timestamp-based methods and the need for decoding streams, leading to misalignment of audio and video data.

Innovation Solution

The proposed technique involves extrapolating signals from delayed audio and video streams to determine synchronization without relying on timestamps, using properties like motion vectors and density vectors to correlate and adjust delays, allowing for synchronization in both compressed and decoded states, thereby reducing computational requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If timestamp-based methods are used for audio-video synchronization, then synchronization can be achieved, but inaccuracies in timestamps lead to misalignment of audio and video data

Engineering Contradiction:
Improvesynchronization accuracyVSAvoidalignment reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces correlation vectors as an intermediary mechanism to bridge audio and video streams. These vectors capture temporal relationships between corresponding frames without relying on timestamps, enabling indirect synchronization that is more reliable against timestamp inaccuracies

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical timestamp-based synchronization system with a signal-processing-based system using correlation vectors. This substitution eliminates the vulnerability to timestamp errors while maintaining synchronization capability through computational correlation analysis

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If existing audio-video synchronization techniques are used, then synchronization can be achieved, but large amounts of computational resources are consumed

Engineering Contradiction:
Improvesynchronization accuracyVSAvoidcomputational resources
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies partial action by computing correlation vectors only at key synchronization points rather than continuously processing entire streams. This selective computation maintains synchronization accuracy while significantly reducing overall computational resource consumption

Inventive Principle:
Principle #16Partial or excessive action

Solution Approach 2:

The patent performs preliminary computation of correlation vectors during encoding or preprocessing stages, so that synchronization can be achieved during playback without requiring intensive real-time computational resources, thus reducing the energy burden during critical synchronization operations

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If streams are decoded for synchronization, then accurate alignment can be achieved, but computational requirements increase and processing time is extended

Engineering Contradiction:
Improvesynchronization accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs correlation vector computation during the encoding or preprocessing phase, before actual playback occurs. This preliminary action allows the synchronization data to be ready in advance, eliminating the need for time-consuming decoding operations during critical synchronization moments

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a simplified representation (correlation vector) of the temporal relationship between audio and video streams that can be processed independently of the actual stream decoding. This copying approach allows synchronization analysis without requiring full stream decoding, reducing processing time

Inventive Principle:
Principle #26Copying

Data Source

PatentUS10063907B1Differential audio-video synchronization
Publication Date: 2018.08.28 HEWLETT PACKARD DEVELOPMENT COMPANY LP
  • US10063907B1 patent drawing
  • US10063907B1 patent drawing
  • US10063907B1 patent drawing

AI summary

Techniques of audio-video synchronization that can be performed on audio and video streams are described. In one scenario, a processor can receive audio-video information comprising a video stream and an audio stream. Each of the video and audio streams can include one or more of: a compressed stream; and a decoded stream. A region of motion in the video stream can be determined based on a property of the video stream. A video signal can be generated based on the determined region of motion. Also, a relative delay between the streams can be determined based, at least in part, on the generated video signal. A delay value associated with one or more of the audio and video streams may be adjusted to assist with synchronizing the audio and video streams.