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 .
DETAILED ACTION
Response to Amendment
This is in response to applicant’s amendment/response filed on 07/24/2026, which has been entered and made of record. Claims 26, 27, 36, 37, 39 have been amended. No claim has been cancelled. No claim has been added. Claims 26-45 are pending in the application.
Response to Arguments
Applicant’s arguments on 07/24/2026 have been fully considered but are moot because the arguments do not apply to any of the references being used in the current rejection.
Claim Objections
Claims 39 are objected to because of the following informalities:
In claim 39, “ssssssssssss.” seems like a typo.
Appropriate correction is required.
Claim Rejections - 35 USC § 103
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 of this title, 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.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim 26-45 are rejected under 35 U.S.C. 103 as being unpatentable over Jourabloo (Jourabloo, Amin, and Xiaoming Liu. "Large-pose face alignment via CNN-based dense 3D model fitting." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.) in view of Avidan et al. (US Pub 2019/0205667 A1).
As to claim 26, Jourabloo discloses a method comprising providing a benchmark 3D model representing a physical object and comprising 3D parameters (Jourabloo, Page 4188, “we propose to use a dense 3D Morphable Model(3DMM) and the projection matrix as the representation of a 2D face image.” Fig. 2, Page 4190, “3.1. 3D Morphable Model” 3DMM representing a head and has parameters.);
varying the 3D parameters within predetermined parameter ranges to generate synthetic 3D objects, wherein each synthetic 3D object comprises corresponding 3D synthetic object parameters (Jourabloo, Fig. 2, Page 4190, Page 4191, “Given a non-frontal 3D face A, we rotate A by using the α and β angles (pitch and yaw angles), and search for a vertex in each defined path which has the maximum (minimum) x coordinate, i.e., the boundary vertex on the right (left) cheek.” The face poses are varying the 3D parameters.);
creating, for each synthetic 3D object, a corresponding synthetic 2D image (Page 4190, “Any 3D face model will be projected onto a 2D image where the face shape may be represented as a sparse set of N landmarks, on the facial fiducial points.”);
calculating, for each synthetic 2D image, corresponding 2D synthetic object parameters (Page 4190, “wheres is a scale parameter, R is the first two rows of a 3×3 rotation matrix controlled by three rotation angles α, β, and γ, t is a translation parameter composed of tx and ty, d is a N-dim index vector indicating the indexes of semantically meaningful 3D vertexes that correspond to 2D landmarks. By collecting all parameters related to this projection, we form a projection vector m = (s,α,β,γ,tx,ty)⊺.” 2D landmarks U are 2D synthetic object parameters.); and
training a neural network to generate 3D object parameters given an input of 2D object parameters based on the 3D synthetic object parameters and the 2D synthetic object parameters (Page 4191, “Specifically, given the labeled visible 2D landmarks U
and the landmark visibilities V, we use the following objective function to estimate m and p,” “which basically minimizes the difference between the projection of 3D landmarks and the 2D labeled landmarks” Page 4191, “Given a set of Nd training face images and their augmented (a.k.a. ground truth in this context) m and p representation, we are interested in learning a mapping function that is able to predict m and p from the appearance of a face image. Clearly this is a complicated non-linear mapping function. Given the success of CNN in vision tasks such as pose estimation [17], face detection [12], and face alignment [34], we decide to marry the CNN with the cascade regressor framework by learning a series of CNN-based regressors to alternate the estimation of m and p. To the best of our knowledge, this is the first time CNN is used in 3D face alignment, with the estimation of over 10 landmarks.” Equation 6 and 7. Page 4192, “Note that since landmark marching is used, the estimated 2D landmarks Ui include the projection of marched 3D landmarks, i.e., 2D cheek landmarks. As a result, the appearance features around these cheek landmarks are part of the input to CNN as well.” Page 4192, “only possible because of the 3D model, can be extracted and contribute to the cascaded CNN learning.”).
Jourabloo does not disclose randomizing the 3D parameters within predetermined parameter ranges to generate synthetic 3D objects.
Avidan discloses randomizing the 3D parameters within predetermined parameter ranges to generate synthetic 3D objects (Avidan, ¶0037, “multiple road geometries and features can be randomly generated by varying a set of randomizable parameters.”. ¶0060, “the image generator 203 can determine one or more randomizable rendering variables associated with the computer-generated image sequence. The image generator then randomizes the one or more randomizable rendering variables to generate the synthetic image data. For example, in order to generate a viable dataset for machine learning (e.g., for CNNs), the video samples included in the synthetic image data may include random changes to several randomizable variables”.).
Jourabloo and Avidan are considered to be analogous art because all pertain to synthetic image. It would have been obvious before the effective filing date of the claimed invention to have modified Jourabloo with the features of “randomizing the 3D parameters within predetermined parameter ranges to generate synthetic 3D objects” as taught by Avidan. The suggestion/motivation would have been randomizing the values of certain parameters enable the machine learning system be trained to be more generalizable with respect to a particular feature (Avidan, ¶0060).
As to claim 27, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses the benchmark 3D model comprises a morphable head model, the predetermined parameter ranges correspond to ranges of human face features and each one of the synthetic 3D objects represents a human head (Jourabloo, Fig. 2, Page 4190, “3.1. 3D Morphable Model”).
As to claim 28, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses calculating, for each synthetic 2D image, corresponding 2D synthetic object parameters comprises extracting a corresponding embedding from each synthetic 2D image (Jourabloo, Page 4190, “wheres is a scale parameter, R is the first two rows of a 3×3 rotation matrix controlled by three rotation angles α, β, and γ, t is a translation parameter composed of tx and ty, d is a N-dim index vector indicating the indexes of semantically meaningful 3D vertexes that correspond to 2D landmarks. By collecting all parameters related to this projection, we form a projection vector m = (s,α,β,γ,tx,ty)⊺.” The landmark is an embedding.).
As to claim 29, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses each synthetic 2D image is created based on a projection of the respective synthetic 3D object (Jourabloo, Page 4190, “Any 3D face model will be projected onto a 2D image where the face shape may be represented as a sparse set of N landmarks, on the facial fiducial points.”).
As to claim 30, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses creating, for each synthetic 3D object, a corresponding parameter pair comprising the corresponding 3D synthetic object parameters and the corresponding 2D synthetic object parameters; and storing the parameter pairs corresponding to the synthetic 3D objects to create a training dataset (Jourabloo, Page 4190, “The relationship between the 3D shape A and 2D landmarks U can be described by using the weak perspective projection” Page 4191, “Given that the projection parameter m and shape parameter p are the representation of a face image, we should have a collection of face images with ground truth m and p so that the learning algorithm can be applied.” “minimizes the difference between the projection of 3D landmarks and the 2D labeled landmarks. Note that although the landmark marching g(:,:) can make cheek landmarks “visible” for non-profile views, the visibility V is useful to avoid invisible landmarks such as outer eye corners and half of the face at the profile view being part of the optimization.”).
As to claim 31, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses for each synthetic 3D object, generating at least one face texture (Jourabloo, Fig. 2, Page 4190, “The core of our proposed 3D face alignment method is the ability to fit a dense 3D Morphable Model to a 2D face image with arbitrary poses.” Fig. 5, Fig. 6, Fig.10-11.).
As to claim 32, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses the 3D object parameters are applicable to the benchmark 3D model for generating a 3D object (Jourabloo, Fig. 2, Fig. 5, Fig. 6, Fig.10-11.).
As to claim 33, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses the 3D object parameters represent deviations with respect to the 3D parameters of the benchmark 3D object (Jourabloo, Fig. 2, Fig. 5, Fig. 6, Fig.10-11. Page 4195, “we estimate the location of landmarks on the cheek, which also drives the 3D face model fitting.”).
As to claim 34, claim 26 is incorporated and the combination of Jourabloo and Avidan discloses the 2D object parameters are calculated from a 2D object image of an object of interest (Jourabloo, Fig. 1, Fig .2, Page 4188, “we propose to use a dense 3D Morphable Model(3DMM) and the projection matrix as the representation of a 2D face image. Therefore, face alignment amounts to estimating this representation, i.e., performing the 3DMMfitting to a face image with arbitrary poses.”).
As to claim 35, claim 34 is incorporated and the combination of Jourabloo and Avidan discloses the 2D object parameters comprise an embedding extracted from the 2D object image (Jourabloo, Page 4188, “employ the powerful cascaded regressor approach to learn the mapping between a 2D face image and its representation.”. Page 4189, “our cascaded CNN can estimate a substantially larger number (34) of 2D and 3D landmarks. Further, using landmark marching [36], our algorithm can adaptively adjust the 3D landmarks during the fitting, so that the cheek landmarks can contribute to the fitting.”).
As to claim 36, the combination of Jourabloo and Avidan discloses a system comprising a training processor configured to: randomize 3D parameters of a benchmark 3D model representing a physical object within predetermined parameter ranges to generate synthetic 3D objects, wherein each synthetic 3D object comprises corresponding 3D synthetic object parameters; create, for each synthetic 3D object, a corresponding synthetic 2D image; calculate, for each synthetic 2D image, corresponding 2D synthetic object parameters; and train a neural network to generate 3D object parameters given an input of 2D object parameters based on the 3D synthetic object parameters and the 2D synthetic object parameters (See claim 26 for detailed analysis.).
As to claim 37, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the benchmark 3D model comprises a morphable head model, the predetermined parameter ranges correspond to ranges of human face features and each one of the synthetic 3D objects represents a human head (See claim 27 for detailed analysis.).
As to claim 38, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the training processor is configured to extract a corresponding embedding from each synthetic 2D image to calculate the corresponding 2D synthetic object parameters (See claim 28 for detailed analysis.).
As to claim 39, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the training processor is configured to create each synthetic 2D image based on a projection of the respective synthetic 3D object (See claim 29 for detailed analysis.).
As to claim 40, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the training processor is configured to create, for each synthetic 3D object, a corresponding parameter pair comprising the corresponding 3D synthetic object parameters and the corresponding 2D synthetic object parameters; and store the parameter pairs corresponding to the synthetic 3D objects to create a training dataset (See claim 30 for detailed analysis.).
As to claim 41, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the training processor is configured to generate, for each synthetic 3D object, at least one face texture (See claim 31 for detailed analysis.).
As to claim 42, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the 3D object parameters are applicable to the benchmark 3D model for generating a 3D object (See claim 32 for detailed analysis.).
As to claim 43, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the 3D object parameters represent deviations with respect to the 3D parameters of the benchmark 3D object (See claim 33 for detailed analysis.).
As to claim 44, claim 36 is incorporated and the combination of Jourabloo and Avidan discloses the 2D object parameters are calculated from a 2D object image of an object of interest (See claim 34 for detailed analysis.).
As to claim 45, claim 44 is incorporated and the combination of Jourabloo and Avidan discloses the 2D object parameters comprise an embedding extracted from the 2D object image (See claim 35 for detailed analysis.).
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to YU CHEN whose telephone number is (571)270-7951. The examiner can normally be reached on M-F 8-5 PST Mid-day flex.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Xiao Wu can be reached on 571-272-7761. 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.
/YU CHEN/
Primary Examiner, Art Unit 2613
Prosecution Insights