DETAILED ACTION
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 .
Response to Amendment
This action is in response to the amendment filed on 20th January, 2026. Claim 21 has been added. Claims 1-21 remain rejected in the application.
Response to Arguments
Applicant's arguments with respect to Claims 1, 8, and 15, filed on 20th January, 2026, with respect to the rejection under 35 U.S.C. § 103 regarding that the prior art does not teach "generate a three-dimensional model of the user’s head based on the set of images, without using any external reference object" and "determine a scaling ratio as a ratio between a distance computed between at least two facial features represented by three-dimensional coordinates within the three-dimensional model of the user's head, and an estimated real-world value associated with said at least two facial features". The proposed arguments have been fully considered, but are not persuasive.
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., “Kornilov-10 lacks metric scaling entirely (it relies on relative classification ratios)” and “Kornilov-10 does not provide a conversion factor to real-world units (e.g., pixels/mm)”) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993). Therefore, applicant’s remark cannot be considered persuasive.
In response to applicant's argument that “Kornilov-10 does not attribute an absolute real-world dimension (e.g., 11.7 mm) to the model,” where “without such a value, the "ratio" in Kornilov-10 cannot be used to "scale" the model to real-world size”, a recitation of the intended use of the claimed invention must result in a structural difference between the claimed invention and the prior art in order to patentably distinguish the claimed invention from the prior art. If the prior art structure is capable of performing the intended use, then it meets the claim. Therefore, applicant’s remark cannot be considered persuasive.
In response to applicant's argument that the prior art does not teach "generate a three-dimensional model of the user’s head based on the set of images, without using any external reference object" as recited in Claim 1, these limitations are taught by Kornilov-10. In particular, Kornilov-10 teaches the following:
Paragraph [0081]: discloses teaches generating a 3D model of the user's face based on reference points obtained from a set of images of said user's face, where the reference points define the locations of various facial features, where no external reference object is used; the reference points are obtained from a set of images of a user's face using only facial features of the user's face as a reference point, which are interpreted as not being an external reference object.
Therefore, applicant’s remark cannot be considered persuasive.
In response to applicant's argument that the prior art does not teach "determine a scaling ratio as a ratio between a distance computed between at least two facial features represented by three-dimensional coordinates within the three-dimensional model of the user's head, and an estimated real-world value associated with said at least two facial features" as recited in Claim 1, these limitations are taught by Kornilov-10. In particular, Kornilov-10 teaches the following:
Paragraph [0100]: discloses using a classifier to determine facial structure/proportions, where "facial proportions may be determined by extracting <read on determining> ratios <read on scaling ratio> between at least two facial feature distances based at least in part on a set of images of the user's face"; and
Paragraph [0083]: discloses the combined reference points <read on estimated real-world value> being 3D coordinates that represent the position of each facial feature location on the 3D surface of the 3D mesh <read on 3D model of user's head>, where the reference points are associated with 3D points of a user's face, such as external eye corners; a ratio will always require at least two quantities; in addition, the reference points are based on points of a set of images, which are being interpreted as approximate or estimated real-world values; furthermore, a classifier uses reference points for its measurements.
Therefore, applicant’s remark cannot be considered persuasive.
In response to applicant's argument that the prior art does not teach "apply the scaling ratio to the three-dimensional model of the user’s head to obtain a scaled user’s head model" as recited in Claim 1, these limitations are taught by the combination of Kornilov-10 and Kornilov-20. In particular, Kornilov-10 teaches the following:
Paragraph [0074]: discloses generating a facial 3D model "by assigning different values <read on applied scaling ratio> to the coefficients associated with each shape or texture component, combining the scaled shape components into a combined face shape, combining the scaled texture components into a combined face texture, and then combining the combined face shape with the combined face texture".
Additionally, Kornilov-20 teaches the following:
Paragraph [0037]: discloses scaling the initial 3D model to obtain an adjusted 3D model <read on scaled model>; a "scaled user's head model" and a "scaled model" are being interpreted as the same terminology.
Therefore, applicant’s remark cannot be considered persuasive.
Regarding arguments to Claims 2-7, 9-14, and 16-21, they directly/indirectly depend on independent Claims 1, 8, and 15 respectively. Applicant does not argue anything other than independent Claims 1, 8, and 15. The limitations in those claims, in conjunction with combination, was previously established as explained.
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, 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, 4-9, 11-13, 15-16, 18-19, and 21 are rejected under 35 U.S.C. 103 as being unpatentable over Kornilov (US 20190164210 A1, previously cited), hereinafter referenced as Kornilov-10, in view of Kornilov et al. (US 20140293220 A1, previously cited), hereinafter referenced as Kornilov-20.
Regarding Claim 1, Kornilov-10 discloses a system (Kornilov-10, [0024]: teaches a system), comprising:
a processor (Kornilov-10, [0024]: teaches the system including a processor); and
a memory coupled to the processor and configured to provide the processor with instructions, which instructions, when executed by the processor, cause the processor to (Kornilov-10, [0024]: teaches the processor being configured to execute instructions stored in memory):
obtain a set of images of a user’s head (Kornilov-10, [0081]: teaches obtaining reference points from a set of images of a user's face);
generate a three-dimensional model of the user’s head based on the set of images, without using any external reference object (Kornilov-10, [0081]: teaches generating a 3D model of the user's face based on reference points obtained from a set of images of said user's face, where the reference points define the locations of various facial features, where no external reference object is used as shown in FIGS. 8A and 8B; Note: it should be noted that the reference points are obtained from a set of images of a user's face using only facial features of the user's face as a reference point, which are interpreted as not being an external reference object);
PNG
media_image1.png
319
397
media_image1.png
Greyscale
detect, from the set of images, a plurality of facial features, including landmarks usable in three dimensions (Kornilov-10, [0081]: teaches obtaining reference points that define facial features from a set of images of the user's face, where each reference point consists of 2D coordinates; [0083]: teaches combining the obtained reference points to create a 3D model of the user's face, where "the corresponding reference points are combined to make a set of (x, y, z) coordinates <read on usable landmarks in 3D> representing the position of each of the reference points (e.g., locations of facial features) on the user's face/head"; Note: it should be noted that it is common in the art for facial recognition applications to include facial landmark extraction, such as the eyes, nose, ears, etc.);
determine a scaling ratio as a ratio between a distance computed between at least two facial features represented by three-dimensional coordinates within the three-dimensional model of the user's head, and an estimated real-world value associated with said at least two facial features (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions, where "facial proportions may be determined by extracting <read on determining> ratios <read on scaling ratio> between at least two facial feature distances based at least in part on a set of images of the user's face"; [0083]: teaches the combined reference points <read on estimated real-world value> being 3D coordinates that represent the position of each facial feature location on the 3D surface of the 3D mesh <read on 3D model of user's head>, where the reference points are associated with 3D points of a user's face, such as external eye corners; Note: it should be noted that a ratio will always require at least two quantities; in addition, the reference points are based on points of a set of images, which are being interpreted as approximate or estimated real-world values; furthermore, a classifier uses reference points for its measurements); and
apply the scaling ratio to the three-dimensional model of the user’s head to obtain a [[scaled]] user’s head model (Kornilov-10, [0074]: teaches generating a facial 3D model "by assigning different values <read on applied scaling ratio> to the coefficients associated with each shape or texture component, combining the scaled shape components into a combined face shape, combining the scaled texture components into a combined face texture, and then combining the combined face shape with the combined face texture").
However, Kornilov-10 does not expressly disclose
apply the scaling ratio to the three-dimensional model of the user’s head to obtain a scaled user’s head model.
Kornilov-20 discloses
apply the scaling ratio to the three-dimensional model of the user’s head to obtain a scaled user’s head model (Kornilov-20, [0037]: teaches scaling the initial 3D model to obtain an adjusted 3D model <read on scaled model>; Note: it should be noted that "scaled user's head model" and "scaled model" are being interpreted as the same terminology).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 8, Kornilov-10 discloses a method for generating a three-dimensional (3D) model (Kornilov-10, [0086]: teaches a method for 3D reconstruction <read on 3D model> of a user's face based on one or more images), comprising:
receiving a set of images of an object (Kornilov-10, [0081]: teaches obtaining reference points from a set of images of a user's face <read on object>);
generating an initial model of the object based on the set of images (Kornilov-10, [0081]: teaches generating a 3D model <read on initial model> of the user's face <read on object> based on reference points obtained from a set of images of said user's face, where the reference points define the locations of various facial features, where no external reference object is used as shown in FIGS. 8A and 8B);
determining a first measurement of a first feature of the object (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions <read on first feature of object> of a user's face, such as facial width and other measurements/ratios of facial features <read on first facial feature>);
classifying the object with a head width classification using a machine learning model based on the set of images (Kornilov-10, [0104]: teaches training <read on machine learning model> a classifier <read on classifying object with head width classification> to identify facial characteristics of a given 3D morphable head model, where a library of images of the user's face is taken from multiple angles);
obtaining, from a storage, a set of proportions corresponding to the head width classification (Kornilov-10, [0029]: teaches the system determining physical characteristics <read on set of proportions> of a user's face for each image, which include facial features such as "3D surface of the face, facial structure or proportion, face shape, eye color, eye shape, hair color, skin color, skin pattern, and the like"; [0044]: teaches images being stored in sensor data storage 316; [0100]: teaches using a classifier <read on head width classification> to determine the facial structure of the user), wherein
the set of proportions comprises an estimated measurement of the first feature of the object associated with the head width classification (Kornilov-10, [0100]: teaches the classifier <read on estimated measurement> receiving facial width and other measures <read on first feature of object associated with head width classification> or ratios of facial features);
determining a scaling ratio for the initial model based on the first measurement and the estimated measurement (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions, where "facial proportions may be determined by extracting <read on determining> ratios <read on scaling ratio> between at least two facial feature distances based at least in part on a set of images of the user's face"; [0100]: further teaches the classifier receiving facial width and other measures <read on estimated measurement> or ratios of facial features); and
scaling the initial model to generate a [[scaled]] model based on the scaling ratio (Kornilov-10, [0074]: teaches generating a facial 3D model "by assigning different values to the coefficients associated with each shape or texture component, combining the scaled shape components into a combined face shape, combining the scaled texture components into a combined face texture, and then combining the combined face shape with the combined face texture").
However, Kornilov-10 does not expressly disclose
scaling the initial model to generate a scaled model based on the scaling ratio.
Kornilov-20 discloses
scaling the initial model to generate a scaled model based on the scaling ratio (Kornilov-20, [0037]: teaches scaling the initial 3D model to obtain an adjusted 3D model <read on scaled model>).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 15, Kornilov-10 discloses a computer program product (Kornilov-10, [0024]: teaches a computer program product),
the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for (Kornilov-10, [0024]: teaches "a computer program product embodied on a computer readable storage medium" and a processor that is "configured to execute instructions stored on and/or provided by a memory coupled to the processor"):
receiving a set of images of a user’s head (Kornilov-10, [0081]: teaches obtaining reference points from a set of images of a user's face);
generating an initial three-dimensional (3D) model of the user’s head based on the set of images, without using any external reference object (Kornilov-10, [0081]: teaches generating a 3D model <read on initial 3D model> of the user's face based on reference points obtained from a set of images of said user's face, where the reference points define the locations of various facial features, where no external reference object is used as shown in FIGS. 8A and 8B);
analyzing the set of images to detect a facial feature on the user’s head (Kornilov-10, [0109]: teaches extracting "facial features <read on detecting facial features> corresponding to a type of facial features based on the received image(s) (1202) <read on set of images>" to be processed <read on analyzing>);
comparing the detected facial feature with an estimated facial feature to determine a scaling ratio (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions <read on detected facial feature>, where "facial proportions may be determined by extracting ratios <read on determine scaling ratio> between at least two facial feature distances based at least in part on a set of images of the user's face"; [0100]: further teaches the classifier receiving facial width <read on estimated facial feature> and other measures/ratios of facial features), wherein
the estimated facial feature comprises at least one of an iris diameter, an ear junction distance, or a temple distance (Kornilov-10, [0059]: teaches facial features <read on estimated facial feature> including "facial structure or proportion, face shape, iris color, hair color, skin color, and skin pattern"); and
scaling the initial 3D model to generate a [[scaled]] 3D model based on the scaling ratio (Kornilov-10, [0074]: teaches generating a facial 3D model "by assigning different values to the coefficients associated with each shape or texture component, combining the scaled shape components into a combined face shape, combining the scaled texture components into a combined face texture, and then combining the combined face shape with the combined face texture").
However, Kornilov-10 does not expressly disclose
scaling the initial 3D model to generate a scaled 3D model based on the scaling ratio.
Kornilov-20 discloses
scaling the initial 3D model to generate a scaled 3D model based on the scaling ratio (Kornilov-20, [0037]: teaches scaling the initial 3D model to obtain an adjusted 3D model <read on scaled model>).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 4, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 1. Kornilov-10 does not expressly disclose the limitations of Claim 4; however, Kornilov-20 discloses wherein the processor is further configured to:
position a glasses frame model on the scaled user’s head model (Kornilov-20, [0032]: teaches "making an initial 3D model of the user's head and determining user head measurements", where "the initial 3D model is used in the process of making an adjusted 3D model <read on scaled user's head>, from which the user's head measurements are determined"; [0027]: teaches "a 3D model of each glasses frame is stored in the database"; [0043]: teaches "the selected glasses are sent to a display to be rendered <read on glasses frame model> on a 3D model of the user's head", where "the 3D model of the user's head is also interactive and the user can interact with the model and see how the glasses may look on the user"); and
determine a set of facial measurements associated with the user’s head based on stored measurement information associated with the glasses frame model (Kornilov-20, [0027]: teaches "a 3D model of each glasses frame is stored in the database", where "other glasses frame information <read on stored measurement information> is stored in the database, including one or more of the following: glasses frame measurements, identifier, name, picture, manufacturer, model number, description, category, type, glasses frame material, brand, part number, and price"; [0028]: teaches using a set of reference points on a user's head 300 to determine user head measurements <read on set of facial measurements>, which are portions of the 3D model of the user's face) and
the position of the glasses frame model on the scaled user’s head model (Kornilov-20, [0024]: teaches rendering selected glasses on a 3D interactive model of the user; [0021]: teaches the initial 3D model of the user's head being adjusted into an adjusted 3D model <read on scaled user's head model>; Note: it should be noted that it would be obvious for one skilled in the art to understand that the rendering of the selected glasses on a 3D model of the user's head would be around the eye area).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 5, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 4. Kornilov-10 does not expressly disclose the limitations of Claim 5; however, Kornilov-20 discloses wherein the processor is further configured to
determine a confidence level corresponding to a facial measurement of the set of facial measurements (Kornilov-20, [0023]: teaches the comparison engine 142 comparing "the user head measurements <read on facial measurements> from the 3D model to a database of glasses frame information 144" using a penalty function, where "one or more glasses frames are selected based on a score <read on confidence level> computed from the penalty function and set thresholds of the score that comprise different levels of fit").
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 6, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 4. Kornilov-10 does not expressly disclose the limitations of Claim 6; however, Kornilov-20 discloses wherein the processor is further configured to:
compare the set of facial measurements to stored dimensions of a set of glasses frames (Kornilov-20, [0027]: teaches "the measurements of the glasses frames comprise a portion of the glasses frame information stored in a database <read on set of glasses frames>"; [0023]: teaches a comparison engine 142, which "compares the user head measurements from the 3D model to a database of glasses frame information 144"); and
output a recommended glasses frame at a user interface based at least in part on the comparison (Kornilov-20, [0042]: teaches a results list that comprises "all glasses in the data that fit the user (i.e., all glasses above the "does not fit" threshold <read on comparison>)", where the user can select one of the resulting glasses <read on recommended glasses frame> from the list; [0043]: teaches displaying <read on user interface> the results list of glasses frames).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 7, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 4. Kornilov-10 does not expressly disclose the limitations of Claim 7; however, Kornilov-20 discloses wherein the processor is further configured to:
input the set of facial measurements into a machine learning model to obtain a set of recommended glasses frames (Kornilov-20, [0042]: teaches a results list <read on set of recommended glasses frame> that comprises "all glasses in the data that fit the user (i.e., all glasses above the "does not fit" threshold)"); and
output the set of recommended glasses frames at a user interface (Kornilov-20, [0042]: teaches a results list <read on recommended glasses frame> that comprises "all glasses in the data that fit the user (i.e., all glasses above the "does not fit" threshold)"; [0043]: teaches displaying <read on user interface> the results list of glasses frames).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 9, the combination of Kornilov-10 and Kornilov-20 discloses the method of Claim 8. Additionally, Kornilov-10 further discloses wherein
the object comprises a user’s head (Kornilov-10, [0100]: teaches a classifier being used to determine the facial structure of the user, which is used to generate a 3D model of the user's face); and
the first feature comprises a face width (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions <read on first feature> of a user's face, such as facial width and other measurements/ratios of facial features).
Regarding Claim 11, the combination of Kornilov-10 and Kornilov-20 discloses the method of Claim 8. Kornilov-10 does not expressly disclose the limitations of Claim 11; however, Kornilov-20 discloses
positioning a 3D model on the scaled model (Kornilov-20, [0032]: teaches "making an initial 3D model of the user's head and determining user head measurements", where "the initial 3D model is used in the process of making an adjusted 3D model <read on scaled model>, from which the user's head measurements are determined"; [0027]: teaches "a 3D model of each glasses frame is stored in the database"; [0043]: teaches "the selected glasses are sent to a display to be rendered <read on 3D model> on a 3D model of the user's head", where "the 3D model of the user's head is also interactive and the user can interact with the model and see how the glasses may look on the user"; Note: it should be noted that "3D model" in this instance is being interpreted as a 3D model representation of glasses), wherein
the 3D model is associated with real-world dimensions (Kornilov-20, [0027]: teaches "measurements of the glasses frames comprise a portion of the glasses frame information stored in a database", where "the glasses frames are scanned with a 3D imager <read on real-world dimensions> and are stored in the database"); and
generating measurements of the object based on the position of the 3D model on the scaled model (Kornilov-20, [0024]: teaches rendering selected glasses <read on 3D model> on a 3D interactive model of the user; [0021]: teaches the initial 3D model of the user's head being adjusted into an adjusted 3D model <read on scaled model>; [0041]: teaches comparing glasses frames <read on position of 3D model> to the user's head measurements <read on object measurements> to generate a fit score using a penalty function) and
a comparison of the 3D model with the scaled model (Kornilov-20, [0027]: teaches "the measurements of the glasses frames comprise a portion of the glasses frame information stored in a database <read on set of glasses frames>"; [0023]: teaches a comparison engine 142, which "compares the user head measurements from the 3D model to a database of glasses frame information 144", where the 3D model of the user's head has already been adjusted).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 12, the combination of Kornilov-10 and Kornilov-20 discloses the method of Claim 8. Kornilov-10 does not expressly disclose the limitations of Claim 12; however, Kornilov-20 discloses
determining measurements of the object based on the scaled model (Kornilov-20, [0032]: teaches adjusting the initial 3D model to obtain an adjusted 3D model <read on scaled model>, "from which the user's head measurements are determined").
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 13, the combination of Kornilov-10 and Kornilov-20 discloses the method of Claim 12. Kornilov-10 does not expressly disclose the limitations of Claim 13; however, Kornilov-20 discloses
determining a confidence level corresponding to each measurement of the measurements (Kornilov-20, [0023]: teaches the comparison engine 142 comparing "the user head measurements <read on measurements> from the 3D model to a database of glasses frame information 144" using a penalty function, where "one or more glasses frames are selected based on a score <read on confidence level> computed from the penalty function and set thresholds of the score that comprise different levels of fit").
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 16, the combination of Kornilov-10 and Kornilov-20 discloses the computer product of Claim 15. Kornilov-10 does not expressly disclose the limitations of Claim 16; however, Kornilov-20 discloses wherein
the estimated facial feature comprises an average measurement of a facial feature in a population (Kornilov-20, [0036]: teaches "the reference points <read on estimated facial feature> from the user's head are averaged <read on average measurement> with other users <read on population> to at least in part generate an initial 3D model"); and
the computer instructions further comprise determining the estimated facial feature (Kornilov-20, [0034]: teaches obtaining reference points <read on estimated facial feature> from images/video frames of different orientations of the user's head).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 18, the combination of Kornilov-10 and Kornilov-20 discloses the computer product of Claim 15. Kornilov-10 does not expressly disclose the limitations of Claim 18; however, Kornilov-20 discloses wherein the computer instructions further comprise:
positioning a 3D model of a glasses frame on the scaled 3D model (Kornilov-20, [0032]: teaches "making an initial 3D model of the user's head and determining user head measurements", where "the initial 3D model is used in the process of making an adjusted 3D model <read on scaled user's head>, from which the user's head measurements are determined"; [0027]: teaches "a 3D model of each glasses frame is stored in the database"; [0043]: teaches "the selected glasses are sent to a display to be rendered <read on glasses frame model> on a 3D model of the user's head", where "the 3D model of the user's head is also interactive and the user can interact with the model and see how the glasses may look on the user"); and
determining facial measurements of the user based on measurements associated with the 3D model of the glasses frame (Kornilov-20, [0027]: teaches "a 3D model of each glasses frame is stored in the database", where "other glasses frame information <read on stored measurement information> is stored in the database, including one or more of the following: glasses frame measurements, identifier, name, picture, manufacturer, model number, description, category, type, glasses frame material, brand, part number, and price"; [0028]: teaches using a set of reference points on a user's head 300 to determine user head measurements <read on set of facial measurements>, which are portions of the 3D model of the user's face) and
the position of the glasses frame on the scaled 3D model (Kornilov-20, [0024]: teaches rendering selected glasses on a 3D interactive model of the user; [0021]: teaches the initial 3D model of the user's head being adjusted into an adjusted 3D model <read on scaled user's head model>; Note: it should be noted that it would be obvious for one skilled in the art to understand that the rendering of the selected glasses on a 3D model of the user's head would be around the eye area).
Kornilov-20 is analogous art with respect to Kornilov-10 because they are from the same field of endeavor, namely analyzing images of user's faces for virtual glasses-fitting. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a neural network that evaluates the fit of glasses on a 3D face model of the user as taught by Kornilov-20 into the teaching of Kornilov-10. The suggestion for doing so would allow the system to train based on weights and penalty functions regardless of facial proportions, thereby improving the accuracy and quality of the glasses-fitting experience. Therefore, it would have been obvious to combine Kornilov-20 with Kornilov-10.
Regarding Claim 19, the combination of Kornilov-10 and Kornilov-20 discloses the computer product of Claim 15. Additionally, Kornilov-10 further discloses wherein the computer instructions further comprise:
determining a head width classification of the user’s head (Kornilov-10, [0100]: teaches a classifier <read on head width classification> receiving facial width and other measures/ratios of facial features to determine the facial structure/facial proportions of the user's face); and
determining the estimated facial feature based on the head width classification of the user’s head (Kornilov-10, [0100]: teaches the classifier receiving facial width <read on estimated facial feature> and other measures/ratios of facial features, where the facial structure/proportions are determined by said classifier).
Regarding Claim 21, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 1. Additionally, Kornilov-10 further discloses wherein the estimated real-world value comprises
a statistical average measurement of said at least two facial features derived from a population (Kornilov-10, [0076]: teaches generating an average face shape <read on statistical average measurement> from various 3D scans of different users' faces 704 <read on population> separately from Principal Component Analysis 706 as shown in FIG. 7, which are based on facial features; [0087]: teaches determining facial features, such as facial structure or proportions; [0100]: teaches facial proportions being determined "by extracting ratios between at least two facial feature distances based at least in part on a set of images of the user's face").
PNG
media_image2.png
152
482
media_image2.png
Greyscale
Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Kornilov (US 20190164210 A1, previously cited), hereinafter referenced as Kornilov-10, in view of Kornilov et al. (US 20140293220 A1, previously cited), hereinafter referenced as Kornilov-20 as applied to Claim 1 above respectively, and further in view of Son et al. (US 20160125228 A1, previously cited), hereinafter referenced as Son.
Regarding Claim 2, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 1. Additionally, Kornilov-10 further discloses wherein the estimated facial features comprise
[[historical facial features, and wherein]]
determining the scaling ratio comprises:determining a measured facial feature from an image of the set of images (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions, where "facial proportions <read on measured facial feature> may be determined by extracting <read on determining> ratios <read on scaling ratio> between at least two facial feature distances based at least in part on a set of images of the user's face");
updating the model of the user’s head based on the measured facial feature (Kornilov-10, [0064]: teaches adjusting the weighting of facial data <read on measured facial data> of a user <read on updating model of user's head>); and
determining the scaling information based on the measured facial feature and at least a portion of the estimated facial features (Kornilov-10, [0074]: teaches generating a 3D model of a face by assigning different values <read on determining scaling information based on measured facial feature> to coefficients associated with each shape or texture component; [0100]: teaches using a classifier <read on estimated facial features> to determine facial structure/proportions).
However, the combination of Kornilov-10 and Kornilov-20 does not expressly disclose
historical facial features.
Son discloses
historical facial features (Son, [0069]: teaches skin analysis information 151 obtaining historical information <read on historical facial features> of the user's face in each face region, where the user can compare past results with other measured results).
Son is analogous art with respect to Kornilov-10, in view of Kornilov-20 because they are from the same field of endeavor, namely image analysis of human faces. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a skin analysis information system to obtain historical information of the user's face as taught by Son into the teaching of Kornilov-10, in view of Kornilov-20. The suggestion for doing so would allow the user to check and compare their face with a past analyzed result. Therefore, it would have been obvious to combine Son with Kornilov-10, in view of Kornilov-20.
Claims 3, 14, 17, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kornilov (US 20190164210 A1, previously cited), hereinafter referenced as Kornilov-10, in view of Kornilov et al. (US 20140293220 A1, previously cited), hereinafter referenced as Kornilov-20 as applied to Claims 1, 8, and 15 above respectively, and further in view of Morrell et al. (US 20230103129 A1, previously cited), hereinafter referenced as Morrell.
Regarding Claim 3, the combination of Kornilov-10 and Kornilov-20 discloses the system of Claim 1. Additionally, Kornilov-10 further discloses wherein determining the scaling ratio comprises:
determining a head width classification corresponding to the user’s head using a machine learning model based on the set of images (Kornilov-10, [0100]: teaches a classifier <read on head width classification> receiving facial width and other measures/ratios of facial features to determine the facial structure/facial proportions of the user's face; [0033]: teaches the recommendation system using a deep neural network <read on using machine learning model>);
[[obtaining a set of proportions corresponding to the head width classification, wherein]]
[[the estimated facial features comprise the set of proportions;]]
determining a measured facial feature from the model of the user’s head (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions <read on measured facial feature> of the user's face); and
determining the scaling ratio based on the measured facial feature and the estimated facial features (Kornilov-10, [0100]: teaches using a classifier to determine facial structure/proportions, where "facial proportions may be determined by extracting <read on determining> ratios <read on scaling ratio> between at least two facial feature distances based at least in part on a set of images of the user's face"; [0100]: further teaches the classifier <read on estimated facial features> receiving facial width and other measures or ratios of facial features).
However, the combination of Kornilov-10 and Kornilov-20 does not expressly disclose
obtaining a set of proportions corresponding to the head width classification, wherein
the estimated facial features comprise the set of proportions.
Morrell discloses
obtaining a set of proportions corresponding to the head width classification (Morrell, [0078]: teaches a landmark detection model 315 being a trained machine learning model that identifies facial geometry <read on head width classification> of the face, where it determines or estimates a variety of facial landmarks <read on set of proportions> based on a single camera input without the need for a dedicated depth sensor), wherein
the estimated facial features comprise the set of proportions (Morrell, [0078]: teaches a landmark detection model 315 being a trained machine learning model that identifies facial geometry of the face, where it determines or estimates a variety of facial landmarks <read on set of proportions of estimated facial features> based on a single camera input without the need for a dedicated depth sensor; [0079]: teaches "the landmark detection model 315 outputs, for each landmark, a set of landmark coordinates").
Morrell is analogous art with respect to Kornilov-10, in view of Kornilov-20 because they are from the same field of endeavor, namely image analysis of human faces. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a landmark detection model to identify facial regions and geometry of the user's face as taught by Morrell into the teaching of Kornilov-10, in view of Kornilov-20. The suggestion for doing so would allow for automatic labelling and classification of facial regions, thereby speeding up 3D model generation. Therefore, it would have been obvious to combine Morrell with Kornilov-10, in view of Kornilov-20.
Regarding Claim 14, the combination of Kornilov-10 and Kornilov-20 discloses the method of Claim 8. The combination of Kornilov-10 and Kornilov-20 does not expressly disclose the limitations of Claim 14; however, Morrell discloses
receiving a second set of images (Morrell, [0081]: teaches the measurement system selecting a set of reference images <read on second set of images> that satisfy an orientation criteria), wherein
each image of the second set of images comprises a learning object including a learning feature associated with a second measurement (Morrell, [0048]: teaches the measurement system determining a second facial height measurement, where "these two measurements can then be compared to determine the relative scale between the images <read on second set of images>"; [0052]: teaches the machine learning system to determine biometric measurements based on captured images; [0078]: teaches the landmark detection model 315 identifying facial geometry of the face, and determining or estimating a variety of facial landmarks <read on learning feature>, where "the landmark detection model 315 may correspond to or include a MediaPipe Face Mesh model <read on learning object>") and
a respective measurement classification (Morrell, [0040]: teaches the measurement system 105 using "a machine learning model to identify and extract, for each image in a stream of images (e.g., in a video), coordinate locations of various facial landmarks (such as the top and bottom of the head, center, top, bottom, and edges of the eyes and mouth, and the like)" in order to evaluate these coordinate locations based on the facial measurements <read on measurement classification>); and
analyzing the second set of images with a machine learning model to associate each respective measurement classification of a set of measurement classifications with a respective second measurement (Morrell, [0052]: teaches "using machine learning to determine biometric (e.g., facial) measurements <read on set of measurement classifications> based on captured images <read on second set of images>"), wherein
the measurement classification is selected from the set of measurement classification to classify the object (Morrell, [0052]: teaches "using machine learning to determine biometric (e.g., facial) measurements <read on measurement classification> based on captured images"; [0071]: teaches an example of the measurement system determining the user's facial height and nose width <read on classify object>).
Morrell is analogous art with respect to Kornilov-10, in view of Kornilov-20 because they are from the same field of endeavor, namely image analysis of human faces. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to a machine learning model to determine biometric measurements of the user's face as taught by Morrell into the teaching of Kornilov-10, in view of Kornilov-20. The suggestion for doing so would result in more accurate landmark measurements of the user's face. Therefore, it would have been obvious to combine Morrell with Kornilov-10, in view of Kornilov-20.
Regarding Claim 17, the combination of Kornilov-10 and Kornilov-20 discloses the computer product of Claim 15. The combination of Kornilov-10 and Kornilov-20 does not expressly disclose the limitations of Claim 17; however, Morrell discloses wherein
the estimated facial feature comprises the iris diameter (Morrell, [0043]: teaches an "iris width 130A and/or 130B (also referred to in some embodiments as iris diameter)"); and
the iris diameter is from 11 mm to 13 mm (Morrell, [0043]: teaches "the average iris diameter or width of an adult human is roughly 11.7 millimeters across a wide population (plus or minus some standard deviation, such as 0.5 millimeters)").
Morrell is analogous art with respect to Kornilov-10, in view of Kornilov-20 because they are from the same field of endeavor, namely facial analysis from different angles. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to define the average iris diameter as taught by Morrell into the teaching of Kornilov-10, in view of Kornilov-20. The suggestion for doing so would allow the system to compare a user's iris to a control, thereby offering eye size information with relation to the average. Therefore, it would have been obvious to combine Morrell with Kornilov-10, in view of Kornilov-20.
Regarding Claim 20, the combination of Kornilov-10 and Kornilov-20 discloses the computer product of Claim 15. The combination of Kornilov-10 and Kornilov-20 does not expressly disclose the limitations of Claim 17; however, Morrell discloses wherein the computer instructions further comprise
associating head width classifications of a set of head width classifications with respective estimated facial features of a set of estimated facial features using a machine learning model that comprises an input of a set of images (Morrell, [0052]: teaches "using machine learning to determine biometric (e.g., facial) measurements <read on set of head width classifications> based on captured images <read on input of set of images>"), wherein
each image of the set of images comprises a head width classification and a facial feature measurement (Morrell, [0052]: teaches "using machine learning to determine biometric (e.g., facial) measurements <read on head width classification> based on captured images <read on input image>"; [0071]: teaches an example of the measurement system determining the user's facial height and nose width in the input image).
Morrell is analogous art with respect to Kornilov-10, in view of Kornilov-20 because they are from the same field of endeavor, namely image analysis of human faces. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to a machine learning model to determine biometric measurements of the user's face as taught by Morrell into the teaching of Kornilov-10, in view of Kornilov-20. The suggestion for doing so would result in more accurate landmark measurements of the user's face. Therefore, it would have been obvious to combine Morrell with Kornilov-10, in view of Kornilov-20.
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Kornilov (US 20190164210 A1, previously cited), hereinafter referenced as Kornilov-10, in view of Kornilov et al. (US 20140293220 A1, previously cited), hereinafter referenced as Kornilov-20 as applied to Claim 9 above respectively, and further in view of Takano et al. (US 20100220933 A1, previously cited), hereinafter referenced as Takano.
Regarding Claim 10, the combination of Kornilov-10 and Kornilov-20 discloses the method of Claim 9. The combination of Kornilov-10 and Kornilov-20 does not expressly disclose the limitations of Claim 10; however, Takano discloses wherein
the head width classification is selected from a list comprising narrow, medium, and wide (Takano, [0164]: teaches a system analyzing a user's face, where it categorizes it into the "short and plump" face type; [0165]: teaches the system analyzing a user's face, where it categorizes it into the "long and refined" face type; FIG. 21 teaches a categorization map, where a user's face type can be determined <read on narrow, medium, and wide>).
PNG
media_image3.png
348
507
media_image3.png
Greyscale
Takano is analogous art with respect to Kornilov-10, in view of Kornilov-20 because they are from the same field of endeavor, namely image analysis of human faces. Before the effective filing date of the claimed invention, it would have been obvious to a person of ordinary skill in the art to implement a categorization map to categorize face types as taught by Takano into the teaching of Kornilov-10, in view of Kornilov-20. The suggestion for doing so would allow for more possible classifications than preset options, thereby offering more classification options for the user. Therefore, it would have been obvious to combine Takano with Kornilov-10, in view of Kornilov-20.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Abitol et al. (US 20030123026 A1) discloses a spectacles fitting system that includes a wide-view imaging system;
Kornilov et al. (US 20180096537 A1) discloses using computed facial feature points to position a product model relative to a model of a face; and
Barton (US 9086582 B1) discloses generating custom-fitted and styled wearable items, such as eyewear, based on measurements made from user-provided image data.
THIS ACTION IS MADE FINAL. 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 nonprovisional extension fee (37 CFR 1.17(a)) 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 mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KARL TRUONG whose telephone number is (703)756-5915. The examiner can normally be reached 10:30 AM - 7:30 PM.
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, Kent Chang can be reached at (571) 272-7667. 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.
/K.D.T./Examiner, Art Unit 2614
/KENT W CHANG/Supervisory Patent Examiner, Art Unit 2614