Model preconditioning for face recognition
AssigneeHID Global Corporation
InventorRobert Kjell Rowe
Life of this patent
Estimating the term…
Abstract
Systems and methods may be used for preconditioning a model, such as for face recognition. The preconditioning may include obtaining a set of facial images, generating, from a plurality of facial images of the set, a plurality of sets of cropped images, each cropped image in the plurality of sets of cropped images including a portion of a face of an image representing a respective set, and preconditioning a machine learning model using the plurality of sets of cropped images. The machine learning model may be refined, such as using a labeled set of captured images of real faces, in an example.
Claims16
A system for comprising:
- a processor;
- memory, including instructions, which when executed by the processor, cause the processor to:
- obtain an unlabeled set of digitally generated facial images;
- generate, from a plurality of facial images of the unlabeled set, a plurality of sets of cropped images, each cropped image in the plurality of sets of cropped images including a portion of a face of an image representing a respective set;
- precondition a machine learning model using the plurality of sets of cropped images, wherein preconditioning the machine learning model includes a first preconditioning that uses images from the plurality of sets of cropped images with overlapping cropping and a second preconditioning that uses only images from the plurality of sets of cropped images that do not include overlapping cropping;
- and
- refine the machine learning model, via transfer learning, using a labeled set of captured images of real faces, the labeled set of captured images alone being insufficient to train a model to achieve a threshold accuracy.
A method for comprising:
- obtaining an unlabeled set of digitally generated facial images;
- using a processor, generating, from a plurality of facial images of the unlabeled set, a plurality of sets of cropped images, each cropped image in the plurality of sets of cropped images including a portion of a face of an image representing a respective set;
- preconditioning, using the processor, a machine learning model using the plurality of sets of cropped images, wherein preconditioning the machine learning model includes a first preconditioning that uses images from the plurality of sets of cropped images with overlapping cropping and a second preconditioning that uses only images from the plurality of sets of cropped images that do not include overlapping cropping;
- and
- refining the machine learning model, via transfer learning, using a labeled set of captured images of real faces, the labeled set of captured images alone being insufficient to train a model to achieve a threshold accuracy.
At least one machine-readable medium, including instructions, which when executed, cause a processor to perform operations comprising:
- obtaining an unlabeled set of digitally generated facial images;
- using a processor, generating, from a plurality of facial images of the unlabeled set, a plurality of sets of cropped images, each cropped image in the plurality of sets of cropped images including a portion of a face of an image representing a respective set;
- preconditioning, using the processor, a machine learning model using the plurality of sets of cropped images, wherein preconditioning the machine learning model includes a first preconditioning that uses images from the plurality of sets of cropped images with overlapping cropping and a second preconditioning that uses only images from the plurality of sets of cropped images that do not include overlapping cropping;
- and
- refining the machine learning model, via transfer learning, using a labeled set of captured images of real faces, the labeled set of captured images alone being insufficient to train a model to achieve a threshold accuracy.
Dependencies are inferred from the claim text.