paper-with-me

PocketNet

2000년 도입 · 논문 2편에서 사용

PocketNet is a face recognition model family discovered through neural architecture search. The training is based on multi-step knowledge distillation.

출처: PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and Multi-Step Knowledge Distillation

소개 논문: PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and Multi-Step Knowledge Distillation

Face Recognition Models · Computer VisionConvolutional Neural Networks · Computer Vision