US7203346B2 — Face recognition method and apparatus using compo…

Claim 1

Independent Claim 139 words

A face recognition apparatus comprising: a component division unit that divides an input facial image into a plurality of facial component images; a face descriptor generating unit that generates face descriptors using a transform matrix corresponding to the respective facial component images, the face descriptors being characteristic vectors, wherein the transform matrix transforms the facial image such that dispersion in the facial image vectors having the same identification decreases and dispersion in the facial image vectors having different identification increases: a registered face descriptor database (DB) that stores registered face descriptors; and an authentication unit that authenticates the input facial image by comparing face descriptors for the input facial image input from the face descriptor generating unit with the registered face descriptors and providing predetermined weights corresponding to each facial component to the comparison results of each facial component.

Claim 2

Dependent Claim 139 words

A face recognition apparatus comprising: a component division unit that divides an input facial image into a plurality of facial component images; a face descriptor generating unit that generates face descriptors using a transform matrix corresponding to the respective facial component images, the face descriptors being characteristic vectors, wherein the transform matrix transforms the facial image such that dispersion in the facial image vectors having the same identification decreases and dispersion in the facial image vectors having different identification increases: a registered face descriptor database (DB) that stores registered face descriptors; and an authentication unit that authenticates the input facial image by comparing face descriptors for the input facial image input from the face descriptor generating unit with the registered face descriptors and providing predetermined weights corresponding to each facial component to the comparison results of each facial component.

Claim 3

Dependent Claim 105 words

The apparatus of claim 1 , further comprising a transform matrix generating unit that generates the transform matrix and outputs the generated transform matrix to the face descriptor generating unit, wherein the transform matrix generating unit comprises: a training image DB that stores training facial images; a training component division unit that divides the training facial images into a plurality of facial component images; and a transform matrix generator that generates a linear transform matrix for each facial component image where the linear transform matrix maximizes a ratio of dispersion in facial images having different identification to dispersion in facial images having the same identification.

Claim 4

Dependent Claim 33 words

The apparatus of claim 1 , wherein the component division unit divides the facial image into a plurality of facial component images such that the divided images are partially overlapped with one another.

Claim 5

Dependent Claim 23 words

The apparatus of claim 1 , wherein the component division unit divides the input facial image into a predetermined number of rectangular images.

Claim 6

Dependent Claim 51 words

The apparatus of claim 1 , wherein the face descriptor generating unit generates face descriptors by decreasing the order of the facial component vectors by performing principal component analysis (PCA) on the facial component images input from the component division unit and transforming the facial component vectors using the transform matrix.

Claim 7

Dependent Claim 77 words

The apparatus of claim 1 , wherein the authentication unit estimates the similarities of the respective facial component images by calculating cross correlation between the face descriptors of the facial component images and the registered face descriptors, providing weights to the similarities in consideration of the facial component images, adding the similarities to which the weights are provided, and authenticating a facial image having the largest addition value to be the same as the input facial image.

Claim 8

Dependent Claim 73 words

The apparatus of claim 1 , wherein the authentication unit estimates the similarities between the face descriptors of the respective facial component images and the registered face descriptors, provides weights to the similarities in consideration of the facial component images, adds the respective similarities to which the weight are provided, and outputs facial images sequentially from a facial image having the highest addition value to a facial image having the lowest addition value.

Claim 9

Dependent Claim 76 words

The apparatus of claim 1 , further comprising a pose estimator that estimates the pose of an image to be recognized by projecting the image to an eigen face which represents predetermined numbers of poses and outputs the image together with information regarding the pose to the component division unit, wherein the component division unit divides the facial image into several parts in consideration of the pose of the facial image input from the pose estimator.

Claim 10

Dependent Claim 116 words

A face recognition method comprising: (a) dividing an input facial image into a plurality of facial component images; (b) generating face descriptors using a transform matrix corresponding to the respective facial components, face descriptors being characteristic vectors representing the respective components of the facial image; wherein the transform matrix transforms the facial image such that dispersion in the facial image vectors having the same identification decreases and dispersion in the facial image vectors having different identification increases; and (c) authenticating the input facial image by comparing the face descriptors of the input facial image with registered face descriptors and providing predetermined weights to the respective comparison results, the predetermined weights corresponding to the respective facial components.

Claim 11

Dependent Claim 116 words

A face recognition method comprising: (a) dividing an input facial image into a plurality of facial component images; (b) generating face descriptors using a transform matrix corresponding to the respective facial components, face descriptors being characteristic vectors representing the respective components of the facial image; wherein the transform matrix transforms the facial image such that dispersion in the facial image vectors having the same identification decreases and dispersion in the facial image vectors having different identification increases; and (c) authenticating the input facial image by comparing the face descriptors of the input facial image with registered face descriptors and providing predetermined weights to the respective comparison results, the predetermined weights corresponding to the respective facial components.

Claim 12

Dependent Claim 72 words

The method of claim 9 , further comprising generating the transform matrix before (a), wherein generating the transform matrix comprises: preparing training image DB that stores training facial images; dividing the training facial images into a plurality of facial component images; and generating a transform matrix for each facial component image, the transform matrix maximizing a ratio of dispersion in facial images having different identification to facial images having the same identification.

Claim 13

Dependent Claim 32 words

The method of claim 9 , wherein in (a) the input facial image is divided into a plurality of facial component images so that the divided images are overlapped with one another.

Claim 14

Dependent Claim 22 words

The method of claim 9 , wherein in (a) the input facial image is divided into a predetermined number of rectangular images.

Claim 15

Dependent Claim 42 words

The method of claim 9 , wherein in (b) the face descriptors are generated by decreasing the orders of facial component vectors by performing PCA on the divided facial component images and by transforming the facial component vectors using the transform matrix.

Claim 16

Dependent Claim 86 words

The method of claim 9 , wherein in (c) the similarities of the respective facial component images are estimated by calculating cross correlation among the face descriptors for the facial component images to be recognized and registered face descriptors, weights are provided to the similarities in consideration of the facial component images, the similarities to which the weights are provided and combined, and a facial image having the largest addition value is authenticated as a facial image to be the same as the input facial image.

Claim 17

Dependent Claim 77 words

The method of claim 9 , wherein during the authentication of the input facial image, the similarities between the respective face descriptors of the respective facial component images, which are divided portions of the facial image, are estimated, weights are provided to the similarities in consideration of the facial component images, the similarities to which the weights are provided and combined, and registered facial images are output sequentially from a facial image having the highest addition value.

Claim 18

Dependent Claim 52 words

The method of claim 9 , before (a) further comprising estimating the pose of the input facial image by projecting the input facial image to eigen faces which represent a predetermined number of poses, wherein in (a) the facial image is divided into facial component images in consideration of the estimated pose.

Claim 19

Dependent Claim 26 words

A recording medium that records the face recognition method claimed in claim 9 as a program code that can be read and executed in a computer.

Claim 20

Dependent Claim 26 words

A recording medium that records the face recognition method claimed in claim 10 as a program code that can be read and executed in a computer.

Claim 21

Dependent Claim 26 words

A recording medium that records the face recognition method claimed in claim 16 as a program code that can be read and executed in a computer.

Claim 22

Independent Claim 153 words

A method of generating a face descriptor for a facial image, comprising: generating first characteristic vectors to represent the face image, using a first transform matrix corresponding to the facial image; dividing the facial image into a plurality of facial component images, and generating second characteristic vectors to represent each of the facial component images, using a second transform matrix corresponding to each of the facial component images; and combining the first characteristic vectors for the facial image and the second characteristic vectors for each of the facial component images to generate the face descriptor, wherein each of the first and second transform matrices transforms vectors of the facial image and the facial component images, such that dispersion in the vectors of the facial image and the facial component images having the same identification decreases and dispersion in the vectors of the facial image and the facial component images having different identification increases.

Claim 23

Independent Claim 153 words

A method of generating a face descriptor for a facial image, comprising: generating first characteristic vectors to represent the face image, using a first transform matrix corresponding to the facial image; dividing the facial image into a plurality of facial component images, and generating second characteristic vectors to represent each of the facial component images, using a second transform matrix corresponding to each of the facial component images; and combining the first characteristic vectors for the facial image and the second characteristic vectors for each of the facial component images to generate the face descriptor, wherein each of the first and second transform matrices transforms vectors of the facial image and the facial component images, such that dispersion in the vectors of the facial image and the facial component images having the same identification decreases and dispersion in the vectors of the facial image and the facial component images having different identification increases.

Claim 24

Independent Claim 164 words

An apparatus of generating a face descriptor for a facial image, comprising: a component division unit that divides the facial image into a plurality of facial component images; and a face descriptor generating unit that generates first characteristic vectors to represent the face image, using a first transform matrix corresponding to the facial image, generates second characteristic vectors to represent each of the facial component images, using a second transform matrix corresponding to each of the facial component images, and combines the first characteristic vectors for the facial image and the second characteristic vectors for each of the facial component images to generate the face descriptor, wherein each of the first and second transform matrices transforms vectors of the facial image and the facial component images, such that dispersion in the vectors of the facial image and the facial component images having the same identification decreases and dispersion in the vectors of the facial image and the facial component images having different identification increases.

Claim 25

Independent Claim 164 words

An apparatus of generating a face descriptor for a facial image, comprising: a component division unit that divides the facial image into a plurality of facial component images; and a face descriptor generating unit that generates first characteristic vectors to represent the face image, using a first transform matrix corresponding to the facial image, generates second characteristic vectors to represent each of the facial component images, using a second transform matrix corresponding to each of the facial component images, and combines the first characteristic vectors for the facial image and the second characteristic vectors for each of the facial component images to generate the face descriptor, wherein each of the first and second transform matrices transforms vectors of the facial image and the facial component images, such that dispersion in the vectors of the facial image and the facial component images having the same identification decreases and dispersion in the vectors of the facial image and the facial component images having different identification increases.

Annuity Stage	Window Opens	Payment Deadline	Surcharge Period Until	Estimated Fee	Status
3.5 Years (Year 4)	2010-04-09	2010-10-09	2011-04-10	$2000.00	Window Closed / Lapsed
7.5 Years (Year 8)	2014-04-09	2014-10-09	2015-04-10	$3760.00	Window Closed / Lapsed
11.5 Years (Year 12)	2018-04-09	2018-10-09	2019-04-10	$7700.00	Window Closed / Lapsed

Face recognition method and apparatus using component-based face descriptor

Abstract

Inventors

Bibliographic Details

Claims Analysis

References Cited (Backward Citations)

USPTO Annuity & Maintenance Windows

Assigned Categories & Tags

Assign Category Tag