Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 5 – 9, 13 – 17 and 20 are rejected under 35 U.S.C 103 as being unpatentable over Hassner US Patent Application Publication No. US-20250342727-A1 (hereinafter Hassner) in view of Wang Patent Application Publication No. CN-117456590-A (hereinafter Wang).
Regarding claim 1, Hassner discloses the method comprising: receiving, from an image capture device, a captured image (Hassner in [0078] discloses, “FIG. 4 shows an identity recognition system 400, such as a face recognition system, when an individual allows to be recognized”); extracting a face image from the captured image (Hassner in [0018] discloses, “a face recognition system would start by detecting the locations of all the faces in the image, often using a face detection component, and obtaining for each face its 2D bounding box coordinates); generating, using a machine learning model and based on the aligned extracted face image, a face embedding (Hassner in [0020] discloses, “Processing continues using a dedicated deep network, often termed an identity embedding network or, in the context of face recognition systems, a face embedding network”); and identifying, based on the generated face embedding and a database of existing face embeddings, an individual associated with the face embedding (Hassner in [0021] discloses, “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize”).
Hassner doesn’t disclose the following limitations as further recited in the claim.
Wang discloses aligning the extracted face image to a reference face model (Wang in [0011] discloses, “step 2.1, face alignment: carrying out standard face feature alignment on the image data in the data set”).
It would have been obvious to one of ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Wang into the system of Hassner because it would allow the system to receive more consistent face image and output more accurate result.
Summary of Citations (Wang)
Paragraph [0011]; “step 2.1, face alignment: carrying out standard face feature alignment on the image data in the data set”.
Summary of Citations (Hassner)
Paragraph [0018]; “a face recognition system would start by detecting the locations of all the faces in the image, often using a face detection component, and obtaining for each face its 2D bounding box coordinates”.
Paragraph [0020]; “Processing continues using a dedicated deep network, often termed an identity embedding network or, in the context of face recognition systems, a face embedding network”.
Paragraph [0021]; “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize”.
Paragraph [0078]; “FIG. 4 shows an identity recognition system 400, such as a face recognition system, when an individual allows to be recognized”.
Regarding claim 5, Hassner in the combination discloses the method of claim 1, wherein determining, based on the generated face embedding and the database of existing face embeddings, the individual associated with the face embedding comprises (Hassner in [0021] discloses, “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize”): for each existing face embedding of the database of existing face embeddings, calculating a similarity score between the generated face embedding and a respective existing face embedding (Hassner in [0056] discloses, “To determine whether the two whole image embedding representations match, the processor can determine the similarity between two multidimensional vectors, which involves calculating a similarity measure”); and comparing the calculated similarity score of between the generated face embedding and the respective existing face embedding to a similarity score threshold value of the respective existing face embedding (Hassner in [0049] discloses, “A match is said to have occurred if the distance between the probe WIER 290 and a gallery item falls below a predetermined threshold. Distances, in this context, can be defined using standard measures of distance between vectors, including L2, cosine, their variants, or alternatives, and the similarity threshold can be determined empirically”).
Summary of Citations (Hassner)
Paragraph [0021]; “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize”.
Paragraph [0049]; “A match is said to have occurred if the distance between the probe WIER 290 and a gallery item falls below a predetermined threshold. Distances, in this context, can be defined using standard measures of distance between vectors, including L2, cosine, their variants, or alternatives, and the similarity threshold can be determined empirically”.
Paragraph [0056]; “To determine whether the two whole image embedding representations match, the processor can determine the similarity between two multidimensional vectors, which involves calculating a similarity measure”.
Regarding claim 6, Hassner discloses the method of claim 1, wherein determining, based on the generated face embedding and the database of existing face embeddings, the individual associated with the face embedding comprises (Hassner in [0021] discloses, “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize”): for each cluster of existing face embeddings within the database of existing face embeddings, obtaining a representative face embedding for a respective cluster of existing face embeddings (Hassner in [0042] discloses about a group of embedding for the same enrolled individual equates to a cluster of embeddings associated with that individual, “In the disclosed system 100, WIER 140 for enrolled individuals can be aggregated into WIER templates capturing appearance information of multiple images previously submitted to the system and labeled as containing the same enrolled individual”); for each representative face embedding, calculating a similarity score between the generated face embedding and the representative face embedding (Hassner in [0056] discloses, “To determine whether the two whole image embedding representations match, the processor can determine the similarity between two multidimensional vectors, which involves calculating a similarity measure”); and comparing the calculated similarity score of between the generated face embedding and the representative face embedding to a similarity score threshold value of the representative face embedding (Hassner in [0049] discloses, “A match is said to have occurred if the distance between the probe WIER 290 and a gallery item falls below a predetermined threshold. Distances, in this context, can be defined using standard measures of distance between vectors, including L2, cosine, their variants, or alternatives, and the similarity threshold can be determined empirically”).
Summary of Citations (Hassner)
Paragraph [0021]; “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize. Gallery representations can be identity embeddings but are often identity templates (similarly, face templates), and each template is an aggregate of the appearance information of a single person as it appears in multiple images or viewing conditions”.
Paragraph [0042]; “In the disclosed system 100, WIER 140 for enrolled individuals can be aggregated into WIER templates capturing appearance information of multiple images previously submitted to the system and labeled as containing the same enrolled individual”.
Paragraph [0049]; “A match is said to have occurred if the distance between the probe WIER 290 and a gallery item falls below a predetermined threshold. Distances, in this context, can be defined using standard measures of distance between vectors, including L2, cosine, their variants, or alternatives, and the similarity threshold can be determined empirically”.
Paragraph [0056]; “To determine whether the two whole image embedding representations match, the processor can determine the similarity between two multidimensional vectors, which involves calculating a similarity measure”.
Regarding claim 7, Hassner in the combination discloses the method of claim 5, wherein each cluster of existing face embeddings in the database of existing face embeddings corresponding to an individual is assigned a unique similarity threshold value (Hassner in [0021] discloses about a template that aggregates multiple appearance of the same person equates to clustering of embedding corresponding to an individual. Furthermore, determining the distance for each person representation and having a predetermined threshold implies to unique similarity threshold value, “Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system ... Gallery representations can be identity embeddings but are often identity templates (similarly, face templates), and each template is an aggregate of the appearance information of a single person as it appears in multiple images or viewing conditions ... The matching process assumes a definition of distances between these representations, such as L2, cosine, their variants, or alternatives. A match occurs if the distance between probe and gallery representations falls below some predetermined threshold, typically determined empirically. Upon matching, the known identity of the matched gallery item is assigned to the probe as the system's recognition result for this detected person in the input visual media.”).
Summary of Citations (Hassner)
Paragraph [0021]; “Identity embeddings serve as probes in a probe-gallery matching system: Identity embeddings are matched against the appearance representations of people who were previously enrolled (stored) in a gallery. The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize. Gallery representations can be identity embeddings but are often identity templates (similarly, face templates), and each template is an aggregate of the appearance information of a single person as it appears in multiple images or viewing conditions ... The matching process assumes a definition of distances between these representations, such as L2, cosine, their variants, or alternatives. A match occurs if the distance between probe and gallery representations falls below some predetermined threshold, typically determined empirically. Upon matching, the known identity of the matched gallery item is assigned to the probe as the system's recognition result for this detected person in the input visual media”.
Regarding claim 8, Hassner discloses the method of claim 1, wherein the database of existing face embeddings includes an embedding of each user at various orientations (Hassner in [0021] discloses about appearance information of a person in multiple viewing condition and [0131] discloses matching under different angles equates to face embeddings includes an embedding of each user at various orientations).
Summary of Citations (Hassner)
Paragraph [0021]; “The gallery is a database, or a subset of a database, containing visual media representations for people known to the system and whom the system is used to recognize. Gallery representations can be identity embeddings but are often identity templates (similarly, face templates), and each template is an aggregate of the appearance information of a single person as it appears in multiple images or viewing conditions”.
Paragraph [0131]; “For 3D models, the recognition process benefits from the additional depth information, allowing for more robust matching even under varying lighting conditions and angles”.
Regarding claim 9, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 1. Therefore, the rejection analysis of claim 1 is applied in claim 9.
Regarding claim 13, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 5. Therefore, the rejection analysis of claim 5 is applied in claim 13.
Regarding claim 14, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 6. Therefore, the rejection analysis of claim 6 is applied in claim 14.
Regarding claim 15, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 7. Therefore, the rejection analysis of claim 7 is applied in claim 15.
Regarding claim 16, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 8. Therefore, the rejection analysis of claim 8 is applied in claim 16.
Regarding claim 17, apparatus claim 17 corresponds to method claim 1. Therefore, the
rejection analysis and motivation to combine of claim 1 is applicable to claim 17.
Regarding claim 20, apparatus claim 20 corresponds to method claim 6. Therefore, the
rejection analysis and motivation to combine of claim 6 is applicable to claim 20.
Claims 2, 10 and 18 are rejected under 35 U.S.C 103 as being unpatentable over Hassner in view of Wang and further in view of Baokanpaku Patent Application Publication No. JP-2019509545-A (hereinafter Baokanpaku).
Regarding claim 2, Wang in the combination disclose the method of claim 1, wherein aligning the extracted face image to a reference face model comprises: identifying a set of landmarks from the extracted face image (Wang in [0106 – 0107] discloses, “Defining a group of standard face feature points, and marking as. Feature points are accurately extracted from an input image using dlib68 point detector”); obtaining, from the set of landmarks, a subset of the set of landmarks; determining, using the subset, the reference face model (Wang in [0104] discloses, “Alignment locates each face by feature point detection (e.g., corner of eyes, nose tip, mouth corners, etc.), and then maps these key feature points onto predetermined standard coordinates using affine transformations”).
The combination of Hassner and Wang doesn’t disclose the following limitations as further recited in the claim.
Baokanpaku discloses a homogenous transformation, parameters of homogenous transformation; and applying the homogenous transformation with the determined parameters to each pixel of the extracted face image (Baokanpaku in [0040] discloses, “for each combination of feature points, the live person's face verification device computes a corresponding transform error using a homographic transform matrix, the transform error being an error of the combination of feature points obtained before and after a homogeneous transform using the homographic transform matrix”).
It would have been obvious to one of ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Baokanpaku into the system of Hassner in view of Wang because it would allow the system to aligned the face more accurately before the face embedding is generated.
Summary of Citations (Baokanpaku)
Paragraph [0040]; “for each combination of feature points, the live person's face verification device computes a corresponding transform error using a homographic transform matrix, the transform error being an error of the combination of feature points obtained before and after a homogeneous transform using the homographic transform matrix; In a case where 50 feature points are preset, a combination of 50 corresponding feature points is obtained, and an average value calculation is performed on a conversion error of the combination of 50 feature points to obtain an average error”.
Summary of Citations (Wang)
Paragraph [0104]; “Alignment locates each face by feature point detection (e.g., corner of eyes, nose tip, mouth corners, etc.), and then maps these key feature points onto predetermined standard coordinates using affine transformations”.
Paragraph [0106 – 0107]; “Defining a group of standard face feature points, and marking as. Feature points are accurately extracted from an input image using dlib68 point detector”.
Regarding claim 10, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 2. Therefore, the rejection analysis of claim 2 is applied in claim 10.
Regarding claim 18, apparatus claim 18 corresponds to method claim 2. Therefore, the
rejection analysis and motivation to combine of claim 2 is applicable to claim 18.
Claims 3 and 11 are rejected under 35 U.S.C 103 as being unpatentable over Hassner in view of Wang and Baokanpaku and further in view of Sun Patent Application Publication No. CN-116704660-A (hereinafter Sun).
Regarding claim 3, Hassner in the combination discloses the method of claim 2.
The combination of Hassner, Wang and Baokanpaku doesn’t disclose the following limitations as further recited in the claim.
Sun discloses the subset comprises a left eye of the face image, a right eye of the face image, and a middle of a lip of the face image (Sun discloses about right, left eye in [0071] and middle of a lip in [0013]).
It would have been obvious to one of ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Sun into the system of Hassner in view of Wang and Baokanpaku because it would allow the system to use feature point of the face to align the non-frontal or angled face image to improve face recognition accuracy.
Summary of Citations (Sun)
Paragraph [0013]; “Dividing the human face into an upper region, a middle region ... the horizontal line of the uppermost position of an upper lip, the middle region is along the horizontal line of the uppermost position of the upper lip to the horizontal line of the topmost position of a nose bridge”.
Paragraph [0071]; ‘if a certain characteristic point is located in the left eye when the user shoots the access control equipment to extract the face, and the left eye is just closed due to dissimilarity, and the right eye is open, the symmetrical characteristic points on the right eye can be used for comparison and identification”.
Regarding claim 11, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 3. Therefore, the rejection analysis of claim 3 is applied in claim 11.
Claims 4, 12 and 19 are rejected under 35 U.S.C 103 as being unpatentable over Hassner in view of Wang and further in view of Nie Patent Application Publication No. CN-107247940-A (hereinafter Nie).
Regarding claim 4, Wang in the combination discloses the method of claim 1, wherein generating the face embedding comprises: inputting, into the machine learning model, the aligned extracted face image to output the face embedding (Wang in [0024] discloses, “by applying the geometric transformation matrix, the whole input image can be adjusted to be aligned with a preset face standard template, so that an aligned face image which is more standardized and easier to identify is obtained”).
The combination of Hassner and Wang doesn’t disclose the following limitations as further recited in the claim.
Nie discloses reducing a dimension of the face embedding (Nie in [0009] discloses, “The invention provides a combined embedding method of face identification and clustering, which avoids an intermediate bottleneck layer, directly optimizes an embedding function, reduces the dimension of a representation vector”).
It would have been obvious to one of ordinary skill in art before the effective filling date of the claimed invention to integrate the technique of Nie into the system of Hassner in view of Wang because reducing dimension makes the face representation more compact so the system can compare the face faster and more efficiently.
Summary of Citations (Nie)
Paragraph [0009]; “The invention provides a combined embedding method of face identification and clustering, which avoids an intermediate bottleneck layer, directly optimizes an embedding function, reduces the dimension of a representation vector, and simplifies the tasks of face verification, face identification and clustering”.
Regarding claim 12, is a non-transitory computer readable storage medium claim
corresponds to apparatus claim 4. Therefore, the rejection analysis of claim 4 is applied in claim 12.
Regarding claim 19, apparatus claim 19 corresponds to method claim 4. Therefore, the
rejection analysis and motivation to combine of claim 4 is applicable to claim 19.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner
should be directed to ZAID MUHAMMAD SALEH whose telephone number is (703)756-1684.
The examiner can normally be reached M-F 8 am - 5 pm ET. Examiner interviews are available
via telephone, in-person, and video conferencing using a USPTO supplied web-based
collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO
Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.If attempts to
reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Vu Le can be
reached on (571)272-7332. The fax phone number for the organization where this application or
proceeding is assigned is 571-273-8300. Information regarding the status of published or
unpublished applications may be obtained from Patent Center. Unpublished application
information in Patent Center is available to registered users. To file and manage patent
submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit
https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center
and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For
additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
If you would like assistance from a USPTO Customer Service Representative, call 800-786
9199 (IN USA OR CANADA) or 571-272-1000.
/ZAID MUHAMMAD SALEH/
Examiner, Art Unit 2668
07/04/2026
/VU LE/Supervisory Patent Examiner, Art Unit 2668
Prosecution Insights