Back to results
US 10,095,917 B2Granted
Systems and methods for facial representation
AssigneeFacebook, Inc.
InventorsYaniv Taigman, Ming Yang, Marc′Aurelio Ranzato
Life of this patent
Estimating the term…
Abstract
Systems, methods, and non-transitory computer readable media can align face images, classify face images, and verify face images by employing a deep neural network (DNN). A 3D-aligned face image can be generated from a 2D face image. An identity of the 2D face image can be classified based on provision of the 3D-aligned face image to the DNN. The identity of the 2D face image can comprise a feature vector.
Claims19
A computer-implemented method, comprising:
- determining, by a computing system, a face portion in a 2D image based at least in part on a first support vector regressor trained to predict point configurations;
- determining, by the computing system, a set of fiducial points in a 2D face image that was generated by transforming the face portion of the 2D image, wherein at least some fiducial points correspond to facial features, and wherein the set of fiducial points are localized based at least in part on a second support vector regressor that is different from the first support vector regressor;
- generating, by the computing system, a 3D-aligned face image from the 2D face image based at least in part on the set of fiducial points detected in the 2D face image and a corresponding set of anchor points placed in a 3D shape of a human face;
- and
- classifying, by the computing system, a facial identity corresponding to the 2D face image based on provision of the 3D-aligned face image to a deep neural network (DNN), wherein the DNN is trained to classify the 2D face image based on a set of features of the 3D-aligned face image.
2 more elements
A system comprising:
- at least one processor;
- and
- a memory storing instructions that, when executed by the at least one processor, cause the system to perform:
- determining a face portion in a 2D image based at least in part on a first support vector regressor trained to predict point configurations;
- determining a set of fiducial points in a 2D face image that was generated by transforming the face portion of the 2D image, wherein at least some fiducial points correspond to facial features, and wherein the set of fiducial points are localized based at least in part on a second support vector regressor that is different from the first support vector regressor;
- generating a 3D-aligned face image from the 2D face image based at least in part on the set of fiducial points detected in the 2D face image and a corresponding set of anchor points placed in a 3D shape of a human face;
- and
- classifying a facial identity corresponding to the 2D face image based on provision of the 3D-aligned face image to a deep neural network (DNN), wherein the DNN is trained to classify the 2D face image based on a set of features of the 3D-aligned face image.
5 more elements
A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform:
- determining a face portion in a 2D image based at least in part on a first support vector regressor trained to predict point configurations;
- determining a set of fiducial points in a 2D face image that was generated by transforming the face portion of the 2D image, wherein at least some fiducial points correspond to facial features, and wherein the set of fiducial points are localized based at least in part on a second support vector regressor that is different from the first support vector regressor;
- generating a 3D-aligned face image from the 2D face image based at least in part on the set of fiducial points detected in the 2D face image and a corresponding set of anchor points placed in a 3D shape of a human face;
- and
- classifying a facial identity corresponding to the 2D face image based on provision of the 3D-aligned face image to a deep neural network (DNN), wherein the DNN is trained to classify the 2D face image based on a set of features of the 3D-aligned face image.
2 more elements
Dependencies are inferred from the claim text.