/faceid

Face Recognition (ML Project)

Primary LanguageJupyter Notebook

Face Identification

The models provide 16 x float32 or 8 x float32 output vectors.
Data distribution allows float32 packing to int16 values, providing 32-byte or 16-byte respective int8 vectors.

Terminology

Face Detection - finding face boundaries on the image.
Face Recognition (Identification) - finding whether two face photos belong to the same person.

Hardware

  • Graphcore IPUs x1/x4/x16
  • Nvidia GPUs x4/x16/x32 (torch: DataParallel, DistributedDataParallel)
  • Google TPUs (no meaningful results)

Credits

TL;DR