Classification system for consumer digital images using workflow, face detection, normalization, and face recognition
First Claim
1. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with an image acquisition device, the programming instructions comprising:
- (a) a face detection module for identifying a group of pixels corresponding to a face region within digital image data acquired by the acquisition device;
(b) a normalization module for generating a normalized face region from said face region, including pose normalizing said face region;
(c) a face recognition module for extracting a set of values of face classifier parameters from said normalized face region, said set of face classifier parameter values being collectively known as a faceprint associated with said normalized face region;
(d) a workflow module for comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, and for determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, the workflow module further for associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints; and
(e) a database module for archiving the data according to the associating by the workflow module within one or more digital data storage media, andwherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing by the workflow module comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, andwherein the determining by the face recognition module comprises automatically determining that the new faceprint corresponds to a known identity based on one or more geometric distance proximities being within a predetermined proximity threshold, andwherein the predetermined proximity threshold comprises a first threshold, and the determining by the workflow module comprises requesting user confirmation whether the normalized face region associated with the new faceprint corresponds to a known identity when a geometric distance proximity is outside the first threshold and within a second threshold greater than the first threshold.
6 Assignments
0 Petitions
Accused Products
Abstract
A processor-based system operating according to digitally-embedded programming instructions includes a face detection module for identifying face regions within digital images. A normalization module generates a normalized version of the face region that is at least pose normalized. A face recognition module extracts a set of face classifier parameter values from the normalized face region that are referred to as a faceprint. A workflow module compares the extracted faceprint to a database of archived faceprints previously determined to correspond to known identities. The workflow module determines based on the comparing whether the new faceprint corresponds to any of the known identities, and associates the new faceprint and normalized face region with a new or known identity within a database. A database module serves to archive data corresponding to the new faceprint and its associated parent image according to the associating by the workflow module within one or more digital data storage media.
-
Citations
155 Claims
-
1. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with an image acquisition device, the programming instructions comprising:
-
(a) a face detection module for identifying a group of pixels corresponding to a face region within digital image data acquired by the acquisition device; (b) a normalization module for generating a normalized face region from said face region, including pose normalizing said face region; (c) a face recognition module for extracting a set of values of face classifier parameters from said normalized face region, said set of face classifier parameter values being collectively known as a faceprint associated with said normalized face region; (d) a workflow module for comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, and for determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, the workflow module further for associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints; and (e) a database module for archiving the data according to the associating by the workflow module within one or more digital data storage media, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing by the workflow module comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, and wherein the determining by the face recognition module comprises automatically determining that the new faceprint corresponds to a known identity based on one or more geometric distance proximities being within a predetermined proximity threshold, and wherein the predetermined proximity threshold comprises a first threshold, and the determining by the workflow module comprises requesting user confirmation whether the normalized face region associated with the new faceprint corresponds to a known identity when a geometric distance proximity is outside the first threshold and within a second threshold greater than the first threshold. - View Dependent Claims (19, 21, 25, 26, 27, 28, 65, 66, 67, 68)
-
-
2. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with a digital image acquisition device the programming instructions comprising:
-
(a) a face detection module for identifying a group of pixels corresponding to a face region within digital image data acquired by the acquisition device; (b) a normalization module for generating a normalized face region from said face region, including pose normalizing said face region; (c) a face recognition module for extracting a set of values of face classifier parameters from said normalized face region, said set of face classifier parameter values being collectively known as a faceprint associated with said normalized face region; (d) a workflow module for comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, and for determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, the workflow module further for associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints; and (e) a database module for archiving the data according to the associating by the workflow module within one or more digital data storage media, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing by the workflow module comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, and wherein the determining by the workflow module comprises requesting user confirmation whether the new faceprint corresponds to a known identity when comparisons of the face classifier parameter values of a first faceprint with multiple archived faceprints corresponding to a same known identity result in at least one determination of an identity match and at least one determination that the identities do not match. - View Dependent Claims (10, 11, 12, 15, 16, 17, 18, 22, 23, 24, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62)
-
-
3. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with a digital image acquisition device, the programming instructions comprising:
-
(a) a face detection module for identifying a group of pixels corresponding to a face region within digital image data acquired by the acquisition device; (b) a normalization module for generating a normalized face region from said face region, including pose normalizing said face region, (c) a face recognition module for extracting a set of values of face classifier parameters from said normalized face region, said set of face classifier parameter values being collectively known as a faceprint associated with said normalized face region; (d) a workflow module for comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, and for determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, the workflow module further for associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints; and (e) a database module for archiving the data according to the associating by the workflow module within one or more digital data storage media, and wherein the set of face classifier parameters comprises a combination of two or more of principle components vectors, independent component vectors, fourier components, discrete cosine transform components, wavelet transform components and gabor transform components, wherein the set of face classifier parameters may be subdivided into two or more subsets of face classifier parameters wherein each subset facilitates a particular step of the comparing and determining a match of said set of face classifier parameters with a previously determined known identity. - View Dependent Claims (63, 64)
-
-
4. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with a digital image acquisition device, the programming instructions comprising:
-
(a) a normalization module for generating a normalized face region from an identified face region within digital image data acquired by the acquisition device, the generating comprising pose normalizing of said face region; and (b) a workflow module providing for the automatic or semiautomatic processing of the pose normalized face region from which face classifier parameter values are extracted and collectively referred to as a faceprint, the processing comprising; (i) comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, (ii) determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, and (iii) associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints, to permit data corresponding to the new faceprint and its associated parent image to be archived according to the associating by the workflow module within one or more digital data storage media, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing by the workflow module comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, and wherein the determining comprises automatically determining that the new faceprint corresponds to a known identity based on one or more geometric distance proximities being within a predetermined proximity threshold, wherein the predetermined proximity threshold comprises a first threshold, and the determining by the workflow module comprises confirming whether the normalized face region associated with the new faceprint corresponds to a known identity when a geometric distance proximity is outside the first threshold and within a second threshold greater than the first threshold. - View Dependent Claims (13, 14, 20, 73, 74, 75, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 109, 110, 111, 112)
-
-
5. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with a digital image acquisition device, the programming instructions comprising:
-
(a) a normalization module for generating a normalized face region from an identified face region within digital image data acquired by the acquisition device, the generating comprising pose normalizing of said face region; and (b) a workflow module providing for the automatic or semiautomatic processing of the pose normalized face region from which face classifier parameter values are extracted and collectively referred to as a faceprint, the processing comprising; (i) comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, (ii) determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, and (iii) associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints, to permit data corresponding to the new faceprint and its associated parent image to be archived according to the associating by the workflow module within one or more digital data storage media, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing by the workflow module comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, and wherein the determining by the workflow module comprises requesting user confirmation whether the new faceprint corresponds to a known identity when comparisons of a face classifier parameter values of the first faceprint with multiple archived faceprints corresponding to a same known identity result in at least one determination of an identity match and at least one determination that the identities do not match. - View Dependent Claims (69, 70, 71, 72, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92)
-
-
6. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with a digital image acquisition device, the programming instructions comprising:
-
(a) a normalization module for generating a normalized face region from an identified face region within digital image data acquired by the acquisition device, the generating comprising pose normalizing of said face region; and (b) a workflow module providing for the automatic or semiautomatic processing of the pose normalized face region from which face classifier parameter values are extracted and collectively referred to as a faceprint, the processing comprising; (i) comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, (ii) determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, and (iii) associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints, to permit data corresponding to the new faceprint and its associated parent image to be archived according to the associating by the workflow module within one or more digital data storage media, and wherein the set of face classifier parameters comprises a combination of two or more of principle components vectors, independent component vectors, fourier components, discrete cosine transform components, wavelet transform components and gabor transform components, wherein the set of face classifier parameters may be subdivided into two or more subsets of face classifier parameters wherein each subset facilitates a particular step of the comparing and determining a match of said set of face classifier parameters with a previously determined known identity. - View Dependent Claims (107, 108)
-
-
7. A processor-based system operating according to digitally-embedded programming instructions residing on one or more processor-readable memories and communicating with one or more digital data storage media for classifying and archiving images including face regions that are acquired with a digital image acquisition device, the programming instructions comprising:
-
(a) a normalization module for generating a normalized face region from an identified face region within digital image data acquired by the acquisition device, the generating comprising; (i) determining the pose of a confirmed face region; (ii) mapping said face region onto a 3d average-face model to create an interim 3d model of said face region; (iii) rotating said interim 3d model into full frontal alignment; and (iv) translating the facial aspect of said interim 3d model back onto 2d image space to create a pose normalized face region; and (b) a workflow module providing for the automatic or semiautomatic processing of the pose normalized face regions, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the workflow module further for determining proximities of values of face classifier parameters of a new face print image with values corresponding to the one or more archived faceprints, and wherein the determining by the workflow module comprises automatically determining that the new face print image corresponds to a known identity based on one or more geometric distance proximities being within a predetermined proximity threshold, wherein the predetermined proximity threshold comprises a first threshold, and the determining by the workflow module comprises confirming whether the normalized face region associated with the new faceprint corresponds to a known identity when a geometric distance proximity is outside the first threshold and within a second threshold greater than the first threshold. - View Dependent Claims (113, 114, 115, 116, 117, 118, 119, 120, 121)
-
-
8. A method for classifying and archiving images including face regions that are acquired with a digital image acquisition device, comprising:
-
using a processor to perform the steps of; (a) generating a pose normalized face region from an identified face region within digital image data acquired by the acquisition device; (b) extracting a set of face classifier parameter values, collectively referred to as a faceprint, from the pose normalized face region; (c) comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, (d) determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, and (e) associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints, to permit data corresponding to the new faceprint and its associated parent image to be archived according to the associating by the workflow module within one or more digital data storage media, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, and wherein the determining comprises automatically determining that the new faceprint corresponds to a known identity based on one or more geometric distance proximities being within a predetermined proximity threshold, and wherein the predetermined proximity threshold comprises a first threshold, and the determining comprises requesting user confirmation whether the normalized face region associated with the new faceprint corresponds to a known identity when a geometric distance proximity is outside the first threshold and within a second threshold greater than the first threshold. - View Dependent Claims (126, 127, 128, 129)
-
-
9. A method for classifying and archiving images including face regions that are acquired with a digital image acquisition device, comprising:
-
using a processor to perform the steps of; (a) generating a pose normalized face region from an identified face region within digital image data acquired by the acquisition device; (b) extracting a set of face classifier parameter values, collectively referred to as a faceprint, from the pose normalized face region; (c) comparing said extracted faceprint to a database of archived faceprints previously determined to correspond to one or more known identities, (d) determining based on the comparing whether a new faceprint corresponds to any of the one or more known identities, and (e) associating the new faceprint and normalized face region from which said faceprint is derived with a new or known identity within a database comprising other data corresponding to the archived faceprints and associated parent images for performing further comparisons with further faceprints, to permit data corresponding to the new faceprint and its associated parent image to be archived according to the associating by the workflow module within one or more digital data storage media, and wherein one or more archived faceprints have been previously determined to correspond to the one or more known identities, and the comparing comprises determining proximities of the values of the face classifier parameters of the new face print image with values corresponding to the one or more archived faceprints, and wherein the determining comprises requesting user confirmation whether the new faceprint corresponds to a known identity when comparisons of the face classifier parameter values of a first faceprint with multiple archived faceprints corresponding to a same known identity result in at least one determination of an identity match and at least one determination that the identities do not match. - View Dependent Claims (122, 123, 124, 125, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155)
-
Specification