facenet-pytorch VS deepface

Hijacking this comment because i've been working nonstop on my project thanks to your suggestion. I'm now using this https://github.com/derronqi/yolov8-face for face detection and still the old face_recognition for encodings. I'm clustering with dbscan and extracting frames with ffmpeg with -hwaccel on. I'm planning to try this: https://github.com/timesler/facenet-pytorch as it looks like it would be the fastest thing avaiable to process videos? Keep in mind i need to perform encoding other than just detection because i want to use DBscan (and later also facial recognition, but this might be done separately just by saving the encodings). let me know if you have any other suggestions, and thanks again for your help

For now, I am trying to evaluate and get the accuracy of the FaceNet module. Like this example on facenet-pytorch, getting the accuracy relies on this file (pairs.txt) provided by the official site. Format description below:

with some python knowledge you can try a two step procedure: 1) extract a number of frames per second, for example five frames (images, i.e. still frames) per second using opencv or ffmpeg 2) Using facenet: detect faces in frames and then classify them by comparing each image to a known image of the person you are looking for.

It's generally better to split the task into a multiple tasks. First I'd want to detect and extract faces. There are a number of pretrained models that you could use for that, e.g. https://github.com/timesler/facenet-pytorch, https://github.com/opencv/opencv/tree/master/data/haarcascades. Once you've extracted faces, you can train a facial recognition using something like a siamese network as you normally would.

You might also want to categorize using either mediapose's holistic model which includes the face mesh or maybe something that gauges emotion like this: https://github.com/serengil/deepface

This is another one I tried but requires programming knowledge. https://github.com/serengil/deepface

It must be a matter of time before we have this, we have so many library's that can detect a face expression. For example: https://github.com/serengil/deepface

I bought a On1 license, but as you point out, it has no facial recognition nor any plans to add it. That also means it has no concept of face regions as implemented in XMP, so that cannot be retrofitted by running face recognition externally, e.g. using an open-source solution like Deepface.

deepface - newer version which has multiple face detectors in it you can mess with, dlib included. Also does face recognition.

If you're looking for models / planning to write some code yourself: Have a look at ArcFace for recognition and RetinaFace for face detection by Insightface. DeepFace repository might also be worth a try.

DeepFace: https://github.com/serengil/deepface

DeepFace utilizes a collection of different face detection models listed here: models = ["VGG-Face", "Facenet", "Facenet512", "OpenFace", "DeepFace", "DeepID", "ArcFace", "Dlib", "SFace"]