Video Call Face Recognition via Shared Media Channel

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing face recognition methods in voice/video calls require installing a dedicated face recognition application on terminals, consuming additional bandwidth and port resources, and often necessitate complex operations and interruptions.

Innovation Solution

Utilize the existing video call media transmission channel to transmit face video streams, eliminating the need for a separate transmission channel and application installation by integrating face recognition into the call process through a media server.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a user equipment simultaneously performs channel estimation and signal decoding, then communication efficiency is improved, but processing complexity and power consumption increase

Engineering Contradiction:
Improvecommunication efficiencyVSAvoidprocessing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent divides the signal processing tasks into separate threads: a first thread dedicated to channel estimation and a second thread dedicated to signal decoding. This segmentation allows each thread to specialize in one function, improving overall efficiency while managing complexity through modular architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs channel estimation before signal decoding by having the first thread complete its estimation task and pass results to the second thread. This preliminary action ensures that decoding can proceed with accurate channel information, improving overall communication efficiency.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If a user equipment simultaneously performs channel estimation and signal decoding, then communication efficiency is improved, but power consumption increases

Engineering Contradiction:
Improvecommunication efficiencyVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent segments the processing workload into separate threads that can execute independently and efficiently. The first thread handles channel estimation while the second thread handles signal decoding, allowing the system to optimize power consumption for each specific task rather than inefficiently processing both simultaneously with a single thread.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

By completing channel estimation as a preliminary action before signal decoding, the system avoids redundant processing and ensures that decoding operations can proceed with pre-computed channel information, reducing overall power consumption compared to simultaneous processing.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If channel estimation is performed before signal decoding, then decoding accuracy is improved, but processing time increases

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

Solution Approach 1:

The patent implements channel estimation as a preliminary action that completes before signal decoding begins. The first thread finishes estimation and passes results to the second thread, ensuring decoding accuracy while the sequential structure manages processing time through efficient task coordination.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

By segmenting the processing into separate threads with the first thread dedicated to estimation and the second to decoding, the system optimizes both accuracy and time management. The segmentation allows each thread to execute its specific task efficiently without resource contention.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4395305B1Communication method, apparatus and system
Publication Date: 2026.05.06 HUAWEI TECH CO LTD
  • EP4395305B1 patent drawingFigure 1
  • EP4395305B1 patent drawingFigure 2
  • EP4395305B1 patent drawingFigure 3

AI summary

A communication method, apparatus, and system are provided, which relate to the field of communication technologies, to reduce bandwidth resources and port resources of a terminal that are occupied by performing face recognition in a call process. The method includes: establishing a video call media transmission channel used to transmit a call video stream between a call terminal and a peer call terminal in a video call service, where the call video stream includes video content photographed by the call terminal or the peer call terminal; receiving, from a media server, a SIP message including a face recognition request identifier, where the face recognition request identifier is for requesting to perform face recognition on a user corresponding to the call terminal; sending a response message of the SIP message to the media server, where the response message of the SIP message indicates that the user corresponding to the call terminal agrees to perform face recognition; next sending a face video stream to the media server through the video call media transmission channel, where the face video stream includes a face image of the user corresponding to the call terminal; and further receiving a face recognition result from the media server.