Scholarlink

This paper seeks to analyze the design and performance enhancement strategies required to support a live surveillance system that incorporates DeepFace for facial identification. With regard to video processing, the system relies on filters from the OpenCV library, which allows for proper face detection and tracking in streaming video. Pysen: DeepFace is linked for precise face recognition so that it does not connect to several faces having the same name. To further improve its speed, multithreading and concurrency approaches are utilized and the resulting system exhibits better reaction time and capability to handle multiple tasks. The gathered results expose a quite efficient system that can perform real-time surveillance with a high enough quality of face recognition to be applicable for security purposes.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Backend Optimization and Face Recognition in Real-Time Surveillance Using DeepFace

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Palivela Thanishka,
  • Pulavarthi Avinash

摘要

This paper seeks to analyze the design and performance enhancement strategies required to support a live surveillance system that incorporates DeepFace for facial identification. With regard to video processing, the system relies on filters from the OpenCV library, which allows for proper face detection and tracking in streaming video. Pysen: DeepFace is linked for precise face recognition so that it does not connect to several faces having the same name. To further improve its speed, multithreading and concurrency approaches are utilized and the resulting system exhibits better reaction time and capability to handle multiple tasks. The gathered results expose a quite efficient system that can perform real-time surveillance with a high enough quality of face recognition to be applicable for security purposes.