Lightweight Real-Time Face Recognition is available on Edge2 and VIM4.
The demo uses lightweight models: RetinaFace and FaceNet. It can detect multiple faces in an image. The fastest recognition speed can reach up to 30 frames.
We have released the Real-Time Face Recognition Demo to help you get started more easily. The only thing you need to do is replace the models with your trained models. The demo provides two similarity comparison methods: Euclidean distance and cosine similarity. You can choose one or both of them to optimize for speed or precision.
Find more details in our documentation.
Doc
Edge2 - Face Recognition Edge2 Demo - 7 [Khadas Docs]
VIM4 - Face Recognition VIM4 Demo - 7 [Khadas Docs]