Back to results
US 11,144,748 B2Granted
Classification system
AssigneeIOT Technology Corporation
InventorKarl L. Wang
Life of this patent
Estimating the term…
Abstract
A biometric classification system includes a biometric capture system that captures a biometric identifier. A classifier includes a plurality of group classifiers. Each group classifier in the plurality of classifiers includes a group discriminator that determines, based on the captured biometric identifier, whether the biometric identifier belongs to a group of persons associated with the group discriminator, and includes a plurality of object discriminators. Each object discriminator is associated with a single person within the group of persons. The group discriminator determines whether the biometric identifier belongs to the group of persons. The object discriminator determines whether the biometric identifier belongs to the single person associated with the object discriminator.
Claims18
An image classification system, comprising:
- a camera system that captures an image of a face;
- an image generator that receives the image and generates a feature-enhanced image;
- and
- a classifier that includes:
- a plurality of group classifiers, each group classifier in the plurality of classifiers including:
- a group discriminator that determines, based on the feature-enhanced image, whether the image of the face belongs to a group of persons associated with the group discriminator, and
- a plurality of object discriminators, each object discriminator being associated with a single person within the group of persons wherein when the group discriminator determines that the image of the face belongs to the group of persons, the object discriminator determines whether the image of the face belongs to the single person associated with the object discriminator, wherein the plurality of object discriminators are on standby and powered down when the group discriminator determines the image of the face does not belong to the group of persons associated with the group discriminator.
4 more elements
An image classification system, comprising:
- a camera system that captures an image of a face;
- an image generator that receives the image and generates a feature-enhanced image;
- and
- a classifier that includes:
- a plurality of group classifiers, each group classifier in the plurality of classifiers including:
- a group discriminator that determines, based on the feature-enhanced image, whether the image of the face belongs to a group of persons associated with the group discriminator, and
- a plurality of object discriminators, each object discriminator being associated with a single person within the group of persons wherein when the group discriminator determines that the image of the face belongs to the group of persons, the object discriminator determines whether the image of the face belongs to the single person associated with the object discriminator, wherein each object discriminator in the plurality of object discriminators is a shallow convolutional neural network consisting of layers trained to classifying multiple images of a single person.
4 more elements
An image classification system, comprising:
- a camera system that captures an image of a face;
- an image generator that receives the image and generates a feature-enhanced image;
- and
- a classifier that includes:
- a plurality of group classifiers, each group classifier in the plurality of classifiers including:
- a group discriminator that determines, based on the feature-enhanced image, whether the image of the face belongs to a group of persons associated with the group discriminator, and
- a plurality of object discriminators, each object discriminator being associated with a single person within the group of persons wherein when the group discriminator determines that the image of the face belongs to the group of persons, the object discriminator determines whether the image of the face belongs to the single person associated with the object discriminator;
- wherein the image generator is trained within a generative adversarial network using a set of facial image data from persons within the group of persons.
5 more elements
A biometric classification system, comprising:
- a biometric capture system that captures a biometric identifier;
- a classifier that includes:
- a plurality of group classifiers, each group classifier in the plurality of classifiers including:
- a group discriminator that determines, based on the biometric identifier captured by the biometric capture system, whether the captured biometric identifier indicates a group of persons associated with the group discriminator, and
- a plurality of object discriminators, each object discriminator being associated with a single person within the group of persons wherein when the group discriminator determines that the captured biometric identifier indicates the group of persons, the object discriminator determines whether the captured biometric identifier indicates the single person associated with the object discriminator;
- and
- a biometric generator that is trained within a generative adversarial network using a set of biometrics from persons within the group of persons, the biometric generator enhancing biometrics captured by the biometric capture system.
4 more elements
An image classification system, comprising:
- a camera system that captures an image of a face;
- an image generator that receives the image and generates a feature-enhanced image;
- and
- a classifier that includes:
- a plurality of group classifiers, each group classifier in the plurality of classifiers including:
- a group discriminator that determines, based on the feature-enhanced image, whether the image of the face belongs to a group of persons associated with the group discriminator, and
- a plurality of object discriminators, each object discriminator being associated with a single person within the group of persons wherein when the group discriminator determines that the image of the face belongs to the group of persons, the object discriminator determines whether the image of the face belongs to the single person associated with the object discriminator;
- wherein the image generator stores a set of model parameters that can be used to generate a set of images, wherein the set of model parameters is numerically smaller than the set of images.
5 more elements
Dependencies are inferred from the claim text.