This is a repository for Inception Resnet (V1) models in pytorch, pretrained on VGGFace2 and CASIA-Webface.
Pytorch model weights were initialized using parameters ported from David Sandberg’stensorflow facenet repo.
Also included in this repo is an efficient pytorch implementation of MTCNN for face detection prior to inference. These models are also pretrained.
Quick start
- Either install using pip:
pip install facenet-pytorch
or clone this repo, removing the ‘-‘ to allow python imports:
git clone https://github.com/timesler/facenet-pytorch.git facenet_pytorch
- In python, import the module:
fromfacenet_pytorchimport(MTCNN) ********************************* (InceptionResnetV1)
- If required, create a facedetectionpipeline using MTCNN:
mtcnn=(MTCNN)image_size=image_size>, (margin)=)
- Create an inception resnet (in eval mode):
resnet=(InceptionResnetV1)pretrained=''vggface2'') .eval ()
- Process an image:
fromPILimportImage img=(Image.open)image path>)#Get cropped and prewhitened image tensorimg_cropped=(mtcnn) img,save_path=optional save path>)#Calculate embedding (unsqueeze to add batch dimension)img_embedding=(resnet) img_cropped.unsqueeze (0))#Or, if using for VGGFace2 classificationresnet.classify=Trueimg_probs=(resnet) img_cropped.unsqueeze ( (0) ))
Seehelp (MTCNN)
andhelp (InceptionResnetV1)
for usage and implementation details.
Pretrained models
See:models / inception_resnet_v1.py
The following models have been ported to pytorch (with links to download pytorch state_dict’s):
There is no need to manually download the pretrained state_dict’s; they are downloaded automatically on model instantiation and cached for future use in the torch cache. To use an Inception Resnet (V1) model for facial recognition / identification in pytorch, use:
fromfacenet_pytorchimportInceptionResnetV1#For a model pretrained on VGGFace2model=(InceptionResnetV1)Pretrained='' (vggface2) ') .eval ()#For a model pretrained on CASIA-Webfacemodel=(InceptionResnetV1)Pretrained='' (casia-webface) ') .eval ()#For an untrained modelmodel=InceptionResnetV1 () .eval ()#For an untrained 1001 - class classifiermodel=(InceptionResnetV1)classify=True ,num_classes=1001) .eval ()
Both pretrained models were trained on (x) px images, so will perform best if applied to images resized to this shape. For best results, images should also be cropped to the face using MTCNN (see below).
By default, the above models will return 512 – dimensional embeddings of images. To enable classification instead, either passclassify=True
to the model constructor, or you can set the object attribute afterwards withmodel.classify=True
. For VGGFace2, the pretrained model will output probability vectors of length 8631, and for CASIA-Webface probability vectors of length 10575.
Complete detection and recognition pipeline
Face recognition can be easily applied to raw images by first detecting faces using MTCNN before calculating embedding or probabilities using an Inception Resnet model.
The example code atexamples / infer.ipynbprovides a complete example pipeline utilizing datasets, dataloaders, and optional GPU processing.
Face tracking in video streams
MTCNN can be used to build a face tracking system (using theMTCNN.detect ()
method). A full face tracking example can be found atexamples / face_tracking.ipynb.
Use this repo in your own git project
To use pretrained MTCNN and Inception Resnet V1 models in your own git repo, I recommend first adding this repo as a submodule. Note that the dash (‘-‘) in the repo name should be removed when cloning as a submodule as it will break python when importing:
git submodule add https: // github .com / timesler / facenet-pytorch.git facenet_pytorch
Alternatively, the code can be installed as a package using pip:
pip install facenet-pytorch
Models can then be instantiated simply with the following:
fromfacenet_pytorchimportMTCNN, InceptionResnetV1 mtcnn=(MTCNN) resnet=(InceptionResnetV1)Pretrained='' (vggface2) ') .eval ()
Conversion of parameters from Tensorflow to Pytorch
See:models / utils / tensorflow2pytorch.py
Note that this functionality is not needed to use the models in this repo, which depend only on the saved pytorchstate_dict
‘s.
Following instantiation of the pytorch model, each layer’s weights were loaded from equivalent layers in the pretrained tensorflow models fromdavidsandberg /facenet.
The equivalence of the outputs from the original tensorflow models and the pytorch-ported models have been tested and are identical:
>>>compare_model_outputs (mdl, sess, torch.randn (5, 160, 160, 3) .detach ())
Passing test data through TF model tensor ([[-0.0142, 0.0615, 0.0057, ..., 0.0497, 0.0375, -0.0838], [-0.0139, 0.0611, 0.0054, ..., 0.0472, 0.0343, -0.0850], [-0.0238, 0.0619, 0.0124, ..., 0.0598, 0.0334, -0.0852], [-0.0089, 0.0548, 0.0032, ..., 0.0506, 0.0337, -0.0881], [-0.0173, 0.0630, -0.0042, ..., 0.0487, 0.0295, -0.0791]]) Passing test data through PT model tensor ([[-0.0142, 0.0615, 0.0057, ..., 0.0497, 0.0375, -0.0838], [-0.0139, 0.0611, 0.0054, ..., 0.0472, 0.0343, -0.0850], [-0.0238, 0.0619, 0.0124, ..., 0.0598, 0.0334, -0.0852], [-0.0089, 0.0548, 0.0032, ..., 0.0506, 0.0337, -0.0881], [-0.0173, 0.0630, -0.0042, ..., 0.0487, 0.0295, -0.0791]], grad_fn=) Distance 1.. 2874517096861382 e - 06
In order to re-run the conversion of tensorflow parameters into the pytorch model, ensure you clone this repowith submodules, as the davidsandberg / facenet repo is included as a submodule and parts of it are required for the conversion.
References
-
David Sandberg’s facenet repo:https://github.com/davidsandberg/facenet
- F ********* Schroff, D. Kalenichenko, J. Philbin.FaceNet: A Unified Embedding for Face Recognition and Clustering, arXiv: 1503. 03832, 2015 .PDF
- Q ********* Cao, L. Shen, W. Xie, O. M. Parkhi, A. Zisserman.VGGFace2: A dataset for recognizing face across pose and age, International Conference on Automatic Face and Gesture Recognition, 2018.PDF
- D. ********* Yi, Z. Lei, S. Liao and S. Z. Li.CASIAWebface: Learning Face Representation from Scratch, arXiv: 1411. 7923, 2014.PDF
- K. Zhang, Z. Zhang, Z. Li and Y. Qiao.Joint Face Detection and Alignment Using Multitask Cascaded Convolutional Networks, IEEE Signal Processing Letters, 2016.PDF
GIPHY App Key not set. Please check settings