Emotion Recognition in Video Conferencing via Facial Mesh Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Video conferencing systems face challenges in detecting and managing negative emotions, such as anger or annoyance, in customers, as these emotions can be difficult for service representatives to recognize, potentially leading to unresolved issues and strained interactions.

Innovation Solution

A computer-implemented method for video conferencing that analyzes facial and speech emotions using convolution neural networks and state vector machines to identify negative emotions, allowing for real-time reporting and intervention by a third party, such as a supervisor, to address customer concerns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If video conferencing is used to provide direct communication with customers, then personalized attention and service quality are improved, but the ability to detect and manage negative emotions becomes more difficult

Engineering Contradiction:
Improveservice qualityVSAvoidemotion detection difficulty
Core Design Contradiction:
Ease of operationVSDifficulty of detecting and measuring

Solution Approach 1:

The patent introduces an emotion recognition system as an intermediary between the customer service representative and the customer. This system automatically analyzes facial expressions, voice tone, and body language to detect negative emotions, serving as a mediator that assists human operators in situations where emotional detection is difficult.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces the mechanical/manual process of emotional detection by human operators with an automated computer vision and audio analysis system. This substitution uses algorithms to process visual and auditory data, identifying negative emotions without requiring human interpretation of subtle emotional cues.

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

2Device complexity

If manual emotion detection by service representatives is used, then system complexity is kept low, but detection accuracy and timeliness deteriorate

Engineering Contradiction:
Improvesystem complexityVSAvoidemotion detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent implements a feedback mechanism where the emotion recognition system continuously monitors the interaction and provides real-time alerts to service representatives when negative emotions are detected. This feedback loop enables timely intervention and allows representatives to adjust their approach based on automated emotional analysis.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The emotion recognition system operates autonomously without requiring manual configuration or intervention. It automatically processes video and audio streams, identifies negative emotions, and generates notifications, enabling the system to serve itself in detecting and reporting emotional states throughout the interaction.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If automated emotion recognition is implemented, then emotion detection accuracy is improved, but system complexity and computational requirements increase

Engineering Contradiction:
Improveemotion detection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the emotion recognition system into distinct functional modules: facial expression analysis, voice tone analysis, body language detection, and integration/decision-making components. Each module processes specific aspects of the interaction independently, then combines results to determine overall emotional state, reducing the complexity of any single component.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3271870B1Emotion recognition in video conferencing
Publication Date: 2023.06.28 SNAP INC
  • EP3271870B1 patent drawingFigure 1A
  • EP3271870B1 patent drawingFigure 1B
  • EP3271870B1 patent drawingFigure 2

AI summary

Methods and systems for videoconferencing include recognition of emotions related to one videoconference participant such as a customer. This ultimately enables another videoconference participant, such as a service provider or supervisor, to handle angry, annoyed, or distressed customers. One example method includes the steps of receiving a video that includes a sequence of images, detecting at least one object of interest (e.g., a face), locating feature reference points of the at least one object of interest, aligning a virtual face mesh to the at least one object of interest based on the feature reference points, finding over the sequence of images at least one deformation of the virtual face mesh that reflect face mimics, determining that the at least one deformation refers to a facial emotion selected from a plurality of reference facial emotions, and generating a communication bearing data associated with the facial emotion.