timesler/facenet-pytorch

Implement alternative detection algorithms

Open

#74 aperta il 16 apr 2020

Vedi su GitHub
 (0 commenti) (2 reazioni) (0 assegnatari)Python (830 fork)batch import
help wanted

Metriche repository

Star
 (3830 star)
Metriche merge PR
 (Nessuna PR mergiata in 30 g)

Descrizione

As highlighted in #10, the current MTCNN design is appropriate for the project while a single detection method is implemented, but with multiple, it makes more sense to separate the return of bounding boxes and confidences from the cropping of images.

Should also think about returning facial landmarks to implement face alignment following detection.

Retinaface is probably the best option to start with.

Guida contributor