Executing Baseline Algorithms¶
In this section we introduce the baselines available in this pakcage. To execute one of then in the databases available just run the following command:
$ bob bio pipeline simple [DATABASE_NAME] [BASELINE]
Note
Both, [DATABASE_NAME] and [BASELINE] can be either python resources or python files.
Please, refer to bob.bio.base for more information.
Baselines available¶
The algorithms below constains all the face recognition baselines available. It is split in two groups, before and after deep learning era.
Before Deep learning era¶
eigenface
: The eigenface algorithm as proposed by [TP91]. It uses the pixels as raw data, and applies a Principal Component Analysis (PCA) on it. Deprecated, please rely on older versions of bob.bio.facelda
: The LDA algorithm (was removed but can be easily implemented) applies a Linear Discriminant Analysis (LDA), here we use the combined PCA+LDA approach [ZKC98] Deprecated, please rely on older versions of bob.bio.facegabor_graph
: This method extract grid graphs of Gabor jets from the images, and computes a Gabor phase based similarity [GHW12]., Deprecated, please rely on older versions of bob.bio.facelgbphs
: Local Gabor binary pattern histogram sequence (LGBPHS) implemented in [ZSG05] Deprecated, please rely on older versions of bob.bio.face
Deep learning baselines¶
facenet-sanderberg
: FaceNet trained by David Sanderberginception-resnetv2-msceleb
: Inception Resnet v2 model trained using the MSCeleb dataset in the context of the work published by [TFP18]inception-resnetv1-msceleb
: Inception Resnet v1 model trained using the MSCeleb dataset in the context of the work published by [TFP18]inception-resnetv2-casiawebface
: Inception Resnet v2 model trained using the Casia Web dataset in the context of the work published by [TFP18]inception-resnetv1-casiawebface
: Inception Resnet v1 model trained using the Casia Web dataset in the context of the work published by [TFP18]arcface-insightface
: Arcface model (Resnet100 backbone) from Insightfaceresnet50-msceleb-arcface-2021
: Resnet Arcface model trained with MSCeleb dataset (dataset partially prunned)resnet50-msceleb-arcface-20210521
: Arcface model trained with MSCeleb dataset (dataset prunned)resnet50-vgg2-arcface-2021
: Arcface model trained with VGG2 datasetiresnet34
: Arcface model (Resnet 34 backbone) from Pytorch InsightFaceiresnet50
: Arcface model (Resnet 50 backbone) from Pytorch InsightFaceiresnet100
: Arcface model (Resnet 100 backbone) from Pytorch InsightFacevgg16-oxford
: VGG16 Face model from Oxfordafffe
: Pytorch network that extracts 1000-dimensional features, trained by Manuel Gunther, as described in [LGB18]