Facial AGE/GENDER PREDICTION sample

  • Description
    Age and gender recognition, along with other demographical profiling, streamline personalized experiences and market targeting. Essential for tailored content, advertising, and societal research.
  • Key Features
    — YOLOV8-Face facial detection model with landmarks
    — NumPy/Numba output postprocessing (box and landmarks)
    — Facial tracking (with Nvidia Tracker)
    — Landmark-based facial alignment with OpenCV-CUDA
    — Age/Gender prediction with a custom model based on MobileNet V2
  • End-To-End Performance
    — 241 FPS on Nvidia A4000
    — 38 FPS on Jetson NX