← Zur Forschung

Forschung & Publikationen

RetinaFace

Single-stage Dense Face Localisation in the Wild

RetinaFace combines accurate face detection with reliable five-point landmark localisation, making it a practical front-end for recognition and alignment-heavy pipelines.

Paper-Details

RetinaFace: Single-stage Dense Face Localisation in the Wild

Veröffentlichung

CVPR 2020

Autor:innen

Jiankang Deng, Jia Guo, Yuxiang Zhou, Jinke Yu, Irene Kotsia, Stefanos Zafeiriou

Paper öffnen

Forschungsüberblick

RetinaFace focuses on detecting difficult faces in real scenes while also returning landmarks that improve alignment quality. That combination makes it useful not only for detection benchmarks, but also for enterprise systems where downstream recognition accuracy depends on consistent cropping and pose normalization.

Produktive Einsatzszenarien

  • Face detection and alignment before recognition
  • Real-time video analytics and entry gate cameras
  • Image ingestion, cropping, and portrait normalization
  • Mobile capture flows that require stable landmark estimation

Kernbeiträge

Jointly predicts face boxes and five facial landmarks in a single-stage detector, reducing pipeline complexity.

Dense supervision and context-aware design improve robustness on small, occluded, blurred, and profile faces.

Provides dependable alignment signals for recognition, quality assessment, and video analytics workflows.

Geschäftlicher Kontakt

Sprechen wir über Ihre geschäftlichen Anforderungen

Kontaktieren Sie InsightFace für Modelllizenzierung, Unterstützung bei Enterprise-Deployments, kundenspezifische KI-Entwicklung oder strategische Partnerschaften.

Wir akzeptieren ausschließlich Anfragen von geschäftlichen E-Mail-Adressen.