Benchmark: Face Detection Using Deep Learning Models and Frameworks

E. Jiménez-Delgado, A. Méndez-Porras, Sixto Enrique Campaña Bastidas, L. Calvo-Valverde, J. Alfaro-Velasco, J. Rodriguez-Salas

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

Resumen

Facial detection has become essential in fields like security and surveillance, requiring efficient and accurate systems. This study evaluates and compares six frameworks—RetinaFace, Mediapipe, YuNet, YoloFace, Haarcascade, and MTCNN—using the WIDER FACE dataset of 3,226 images. Performance is assessed on an Intel Core i5 13th generation 13600K CPU with 32 GB RAM, focusing on faces recognized, processing time, and average detection time per face. RetinaFace demonstrated the highest detection capability with 22,738 faces but required significant processing time (23,153.78 s, 3.83 s/face). In contrast, YuNet was the most efficient, averaging 0.03 s per face, followed by Mediapipe at 0.04 s, although with fewer detections. MTCNN and YoloFace showed intermediate performance in detection and efficiency. This study highlights key differences in deep learning-based facial detection frameworks, offering insights for future research and applications in the field.

Idioma originalInglés
Título de la publicación alojadaInformation Technology and Systems - ICITS 2025
EditoresAlvaro Rocha, Carlos Ferrás, Hiram Calvo
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas104-113
Número de páginas10
ISBN (versión impresa)9783031931024
DOI
EstadoPublicada - 2025
EventoInternational Conference on Information Technology and Systems, ICITS 2025 - Mexico City, México
Duración: 22 ene 202525 ene 2025

Serie de la publicación

NombreLecture Notes in Networks and Systems
Volumen1449 LNNS
ISSN (versión impresa)2367-3370
ISSN (versión digital)2367-3389

Conferencia

ConferenciaInternational Conference on Information Technology and Systems, ICITS 2025
País/TerritorioMéxico
CiudadMexico City
Período22/01/2525/01/25

Huella

Profundice en los temas de investigación de 'Benchmark: Face Detection Using Deep Learning Models and Frameworks'. En conjunto forman una huella única.

Citar esto