Work

Comparative Analysis of Computer Vision Algorithms for Detection of Emotional Burnout Markers

Year: 2026

Type: article

Abstract: This article examines the problem of automatically detecting signs of emotional burnout in employees working remotely or in an office environment. A comparative analysis of three computer vision algorithms - Dlib, DeepFace (based on convolutional neural networks), and MediaPipe - is conducted based on the accuracy of facial biometric feature extraction, computational load, applicability under background monitoring, and the ability to detect facial microdynamics. Based on these characteristics, the MediaPipe framework is recognized as the most promising solution, enabling the construction of a 3D facial mesh from 478 points and the calculation of 52 coefficients with minimal computational load.

Source: Applied Mathematics and Fundamental Informatics

DOI: 10.25206/2311-4908-2026-13-2-36-41

Authors: N. A. MOISEEVA, M. M. KHOLOD

Institution: Omsk State Technical University

Language: English

Topic: Face recognition and analysis

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: diamond

OpenAlex ID: W7167680903

FWCI:

Cites:

Cited by:

Related to: 10