DETAILED ACTION
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA . The preliminary Amendment filed 07 May 2025 has been entered and considered. Claim 24 has been added. Claims 1-24 are all the claims pending in the application.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 10/14/2025 was considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 15 and 23 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter. In particular, Claims 15 and 23 are directed to a computer storage medium. The United States Patent and Trademark Office (USPTO) is obliged to give claims their broadest reasonable interpretation consistent with the specification during proceedings before the USPTO. See In re Zletz, 893F.2d 319 (Fed. Cir. 1989) (during patent examination the pending claims must be interpreted as broadly as their terms reasonably allow). The broadest reasonable interpretation of Claims 15 and 23 drawn to a computer storage medium covers forms of non-transitory tangible media and transitory propagating signals per se in view of the ordinary and customary meaning of computer readable media. See MPEP 2111.01. When the broadest reasonable interpretation of a claim covers a signal per se, the claim must be rejected under 35 U.S.C. § 101 as covering non-statutory subject matter. See In re Nuijten, 500 F.3d 1346, 1356-57 (Fed. Cir. 2007) (transitory embodiments are not directed to statutory subject matter). Applicants are advised to amend Claims 15 and 23 to recite “A non-transitory computer-readable storage medium…” in order to overcome the rejection.
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.
The factual inquiries 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.
Claims 1-17 and 22-23 are rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (U.S. Patent Pub No. 6,151,032 A, hereafter referred as Cheng) in view of Wang et al. (NPL: DeepFont: Identify Your Font from An Image, hereafter referred as Wang).
Regarding Claim 1:
Cheng teaches a computer-implemented method comprising (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.): determining characterization data for a set of character glyphs of a first font, wherein the characterization data represents one or more stroke attributes indicative of using numerical control to render each stroke of the character glyph (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke.).
Cheng fails to teach and using the characterization data to inform available font options for font selection.
Wang, like Cheng, is directed to image processing techniques for font analysis. Wang does teach and using the characterization data to inform available font options for font selection (Wang: p. 458; The authors in [14] proposed to select fonts based on online crowdsourced attributes, and explore font similarity, from which a user is enabled to explore other visually similar fonts given a specific font. The font similarity measure is very helpful for font selection, organization, browsing, and suggestion.).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cheng to utilize the font detection technique, as taught by Wang, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods for improvement to yield predictable results. As taught by Wang, the proposed modification is helpful for font selection, organization, browsing, and suggestion (Wang: p. 458).
In regards to Claim 2, Cheng as modified by Wang further teaches the computer-implemented method of claim 1, further comprising: determining characterization data for a plurality of fonts (Cheng: Par. (26); In the first step of the process in FIG. 2A, at block 302, a font designer retrieves a previously generated set of stroke-based characters. The set of stroke-based characters are preferably the same font type. If the stroke-based characters are Chinese characters, the following are examples of different Chinese character fonts a designer may choose from: text hand; running hand; cursive hand; seal type; Sung-type; and Ming-type.); and generating a font genome by aggregating the determined characterization data for the plurality of fonts and the determined characterization for the set of character glyphs of the first font (Wang: p. 455; more levels of representation can give rise to more abstract, more general features of the raw input, and that the lower layers of the predictor constitute a hierarchy of features that can be shared across variants of the input distribution. The authors in [7] used data from the union of all domains to learn their shared features, which is different from many previous domain adaptation methods that focus on learning features in a unsupervised way from the target domain only.).
In regards to Claim 3, Cheng as modified by Wang further teaches the computer-implemented method of claim 1, wherein determining characterization data comprises: defining one or more keypoints for each character glyph; and characterizing one or more strokes by determining a set of attributes for each keypoint using an image classification model (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke; obvious to one skilled in the art that the labeling of the key points by the user from Cheng can be implemented into the classification model of Wang).
In regards to Claim 4, Cheng as modified by Wang further teaches the computer-implemented method of claim 3, wherein defining the one or more keypoints for each character glyph comprises processing each character glyph using an image correspondence model to determine the one or more keypoints (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke; obvious to one skilled in the art that the labeling of the key points by the user from Cheng can be implemented into the classification model of Wang).
In regards to Claim 5, Cheng as modified by Wang further teaches the computer-implemented method of claim 4, wherein the one or more keypoints determined by the image correspondence model are generalizable (Cheng: Par. (5); The steps shown in FIG. 2A illustrate the process for generating basic outline glyphs from predefined stroke-based characters. The predefined stroke-based characters are defined in character space by predefined key points and the key points width value at multiple levels of resolution.).
In regards to Claim 6, Cheng as modified by Wang further teaches the computer-implemented method of claim 3, wherein defining the one or more keypoints for each character glyph comprises: identifying a bounding box relative to a cardinal direction on the character glyph; advancing in a direction from a selected point along a contour of each stroke in a set of strokes in the bounding box until one or more criteria are met; and defining the one or more key points in accordance with the criteria each contour of each stroke meets in the bounding box (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke; obvious to one skilled in the art that the labeling of the key points by the user from Cheng can be implemented into the classification model of Wang by utilizing bounding boxes).
In regards to Claim 7, Cheng as modified by Wang further teaches the computer-implemented method of claim 6, wherein the one or more criteria met comprise one or more of width, angle, or rate of change criteria (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke).
In regards to Claim 8, Cheng as modified by Wang further teaches the computer-implemented method of claim 6, further comprising measuring an angle of each contour of each stroke relative to a closest perpendicular stroke to the contour (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke).
In regards to Claim 9, Cheng as modified by Wang further teaches the computer-implemented method of claim 3, further comprising grouping character glyphs into logical groups based on the determined stroke attributes (Cheng: Par. (27); Next, at block 304, a group of similar glyphs are identified from the set of stroke-based characters. The step of identifying is as simple as a font designer visually scanning the set of stroke-based characters, or automatically scanning the set of stroke-based characters by image analysis techniques for glyphs with similar topography. Then, at block 306, the most commonly appearing glyph of a group of glyphs that have a similar topography is selected. This step can also be performed manually or automatically.).
In regards to Claim 10, Cheng as modified by Wang further teaches the computer-implemented method of claim 3, further comprising evaluating a measure of distance between the defined one or more keypoints (Cheng: Par. (42-43); A stroke may not be moved up or down if it fails to fulfill a predefined criteria. In the example of FIGS. 12A and B, the criteria is that the horizontal strokes must maintain at least a one pixel distance from the end of vertical stroke 426 of which they make contact. FIG. 12B illustrates the displayed result of the procedure described above, As shown in FIG. 13A, vertical strokes 430, 432 and 434 are displayed in the character shown with the center stroke 432 one pixel distance closer to right vertical stroke 434 than to left stroke 430. However, in this particular character, the vertical strokes 430, 432 and 434 must maintain symmetry (equal white space between the strokes). The following procedures are invoked if this situation occurs. First, the character is checked to see if the smaller white space can expand. If the smaller space between the strokes cannot be expanded, the larger space between the strokes is reduced to equal that of the smaller space. In the example of FIG. 13A, right stroke 434 is moved one space to the right to make the white space between it and the center stroke 432 equal to two pixels, thus equal to the pixel space between center stroke 432 and left stroke 430. If right stroke 434 had no room to move, left stroke 430 would have moved one pixel closer to center pixel 432, thus making the spaces between the strokes equal.).
In regards to Claim 11, Cheng as modified by Wang further teaches the computer-implemented method of claim 1, wherein using the characterization data to inform available font options for font selection further comprises using the characterization data to inform search engine results (Wang: p. 458; The authors in [14] proposed to select fonts based on online crowdsourced attributes, and explore font similarity, from which a user is enabled to explore other visually similar fonts given a specific font. The font similarity measure is very helpful for font selection, organization, browsing, and suggestion, p. 451; Several websites allow users to search and recognize fonts by font similarity, including Identifont, MyFonts, WhatTheFont, and Fontspring.).
In regards to Claim 12, Cheng as modified by Wang further teaches the computer-implemented method of claim 1, wherein using the characterization data to inform available font options for font selection further comprises conditioning a machine learning model for font selection using embeddings of the characterization data (Wang: p. 457; We implemented and evaluated the local feature embedding based algorithm (LFE) in [4] as a baseline, and include the four different DeepFont models as specified in Table 5.).
In regards to Claim 13, Cheng as modified by Wang further teaches the method of claim 12, wherein using the characterization data to inform available font options for font selection further comprises conditioning a generative machine learning model to generate fonts using the embeddings of the characterization data (Wang: p. 457; We implemented and evaluated the local feature embedding based algorithm (LFE) in [4] as a baseline, and include the four different DeepFont models as specified in Table 5, p. 453; To create a sufficiently large set of synthetic training data, we follow the same way in [4] to render long English words sampled from a large corpus, and generate tightly cropped, gray-scale, and size-normalized text images.).
Regarding Claim 14:
Cheng as modified by Wang further teaches a system comprising one or more computer and one or more storage devices storing instructions that are operable, when executed by the one or more computers (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.), to cause the one or more computers to perform operations comprising: determining characterization data for a set of character glyphs of a first font, wherein the characterization data represents one or more stroke attributes indicative of using numerical control to render each stroke of the character glyph (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke.); and using the characterization data to inform available font options for font selection (Wang: p. 458; The authors in [14] proposed to select fonts based on online crowdsourced attributes, and explore font similarity, from which a user is enabled to explore other visually similar fonts given a specific font. The font similarity measure is very helpful for font selection, organization, browsing, and suggestion.).
Regarding Claim 15:
Cheng as modified by Wang further teaches a computer storage medium encoded with a computer program, the program comprising instructions that are operable, when executed by a data processing apparatus (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.), to cause the data processing apparatus to perform operations comprising: determining characterization data for a set of character glyphs of a first font, wherein the characterization data represents one or more stroke attributes indicative of using numerical control to render each stroke of the character glyph (Cheng: Par. (27); A11 intersecting points for all overlapping strokes in the selected basic glyph are determined. See block 308. Intersections between strokes can be determined by manually selecting the points at which the outlines of the strokes intersect or can be automatically determined using an intersecting algorithm. Since outlines of strokes are predefined as Bezier curves, the intersecting algorithm can easily determine the curve or subcurve segments of a first stroke that intersect with the curve or subcurve segments of a second stroke within the glyph, Par. (39); In order to maintain consistency for like-widthed strokes, a stroke's width dimension and number of pixels is first determined. If the determination results in an odd pixel value, the key points of the stroke are positioned to the center of the pixel in which they are initially located, otherwise, the key point is adjusted to the nearest grid line (line separating pixels) parallel to the lengthwise dimension of the stroke.); and using the characterization data to inform available font options for font selection (Wang: p. 458; The authors in [14] proposed to select fonts based on online crowdsourced attributes, and explore font similarity, from which a user is enabled to explore other visually similar fonts given a specific font. The font similarity measure is very helpful for font selection, organization, browsing, and suggestion.).
Regarding Claim 16:
Cheng as modified by Wang further teaches a computer-implemented method comprising (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.): receiving font genome data for a subset of character glyphs in a font, wherein the font genome data comprises characterization data for each character glyph in the subset (Wang: p. 455; more levels of representation can give rise to more abstract, more general features of the raw input, and that the lower layers of the predictor constitute a hierarchy of features that can be shared across variants of the input distribution. The authors in [7] used data from the union of all domains to learn their shared features, which is different from many previous domain adaptation methods that focus on learning features in a unsupervised way from the target domain only.); conditioning a generative machine learning model to generate fonts using the font genome data for the subset of character glyphs in the font; and generating a font comprising a set of character glyphs not in the subset of character glyphs in the font (Wang: p. 452-454; One way to overcome the training data challenge is to synthesize the training set by rendering text fragments for all the necessary fonts, To create a sufficiently large set of synthetic training data, we follow the same way in [4] to render long English words sampled from a large corpus, and generate tightly cropped, gray-scale, and size-normalized text images.).
In regards to Claim 17, Cheng as modified by Wang further teaches the computer-implemented method of claim 16, wherein conditioning the generative machine learning model to generate fonts using the font genome data comprises using embeddings of the characterization data (Wang: p. 457; We implemented and evaluated the local feature embedding based algorithm (LFE) in [4] as a baseline, and include the four different DeepFont models as specified in Table 5.).
Regarding Claim 22:
Cheng as modified by Wang further teaches a system comprising one or more computer and one or more storage devices storing instructions that are operable, when executed by the one or more computers (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.), to cause the one or more computers to perform operations comprising: receiving font genome data for a subset of character glyphs in a font, wherein the font genome data comprises characterization data for each character glyph in the subset (Wang: p. 455; more levels of representation can give rise to more abstract, more general features of the raw input, and that the lower layers of the predictor constitute a hierarchy of features that can be shared across variants of the input distribution. The authors in [7] used data from the union of all domains to learn their shared features, which is different from many previous domain adaptation methods that focus on learning features in a unsupervised way from the target domain only.); conditioning a generative machine learning model to generate fonts using the font genome data for the subset of character glyphs in the font; and generating a font comprising a set of character glyphs not in the subset of character glyphs in the font (Wang: p. 452-454; One way to overcome the training data challenge is to synthesize the training set by rendering text fragments for all the necessary fonts, To create a sufficiently large set of synthetic training data, we follow the same way in [4] to render long English words sampled from a large corpus, and generate tightly cropped, gray-scale, and size-normalized text images.).
Regarding Claim 23:
Cheng as modified by Wang further teaches a computer storage medium encoded with a computer program, the program comprising instructions that are operable, when executed by a data processing apparatus (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.), to cause the data processing apparatus to perform operations comprising: receiving font genome data for a subset of character glyphs in a font, wherein the font genome data comprises characterization data for each character glyph in the subset (Wang: p. 455; more levels of representation can give rise to more abstract, more general features of the raw input, and that the lower layers of the predictor constitute a hierarchy of features that can be shared across variants of the input distribution. The authors in [7] used data from the union of all domains to learn their shared features, which is different from many previous domain adaptation methods that focus on learning features in a unsupervised way from the target domain only.); conditioning a generative machine learning model to generate fonts using the font genome data for the subset of character glyphs in the font; and generating a font comprising a set of character glyphs not in the subset of character glyphs in the font (Wang: p. 452-454; One way to overcome the training data challenge is to synthesize the training set by rendering text fragments for all the necessary fonts, To create a sufficiently large set of synthetic training data, we follow the same way in [4] to render long English words sampled from a large corpus, and generate tightly cropped, gray-scale, and size-normalized text images.).
Claims 18-21 and 24 are rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (U.S. Patent Pub No. 6,151,032 A, hereafter referred as Cheng) in view of Wang et al. (NPL: DeepFont: Identify Your Font from An Image, hereafter referred as Wang) and He et al. (NPL: Diff-Font: Diffusion Model for Robust One-Shot Font Generation, hereafter referred as He).
In regards to Claim 18, Cheng as modified by Wang fails to further teach the computer-implemented method of claim 17, wherein the generative machine learning model comprises a stable diffusion model, and wherein conditioning the stable diffusion model using the embeddings of the characterization data further comprises: integrating the characterization embeddings into an embedding layer of the stable diffusion.
He, like Cheng, is directed to image processing techniques for font analysis. He does teach wherein the generative machine learning model comprises a stable diffusion model (He: p. 3; We present Diff-Font, a unified generative network for robust one-shot font generation based on the diffusion model. In comparison to GAN-based methods, Diff-Font offers the advantages of stable training and the ability to be effectively trained on large datasets. To the best of our knowledge, this is the first attempt to develop a diffusion model for font generation.), and wherein conditioning the stable diffusion model using the embeddings of the characterization data further comprises: integrating the characterization embeddings into an embedding layer of the stable diffusion model (He: p. 5; Similar to word embedding in the NLP community, we adopt an embedding layer to convert different tokens of characters into different content representations.).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cheng to utilize the diffusion model, as taught by He, to arrive at the claimed invention discussed above. Such a modification is the result of combining prior art elements according to known methods for improvement to yield predictable results. As taught by He, the proposed modification outperforms prior font generation methods in both high-fidelity font style replication and the generation of intricate characters (He: p. 1).
In regards to Claim 19, Cheng as modified by Wang and He further teaches the computer-implemented method of claim 18, wherein conditioning the generative machine learning model to generate fonts using the font genome data further comprises receiving one or more control parameters indicative of desired font characteristics (He: p. 5; ControlNet Zhang and Agrawala (2023) introduces an extra conditional control module to enable a pre-trained diffusion model to be applied to specific tasks. This work is further extended by the multi-attribute conditional diffusion model which introduces composite-wise and stroke-wise attributes conditional for better training and attribute-wise diffusion guidance strategy for stroke-aware or component-aware font generation.).
In regards to Claim 20, Cheng as modified by Wang and He further teaches the computer-implemented method of claim 19, wherein the subset of character glyphs in the font is representative of a mutually exclusive subset of character glyphs in the font (Wang: p. 452-454; One way to overcome the training data challenge is to synthesize the training set by rendering text fragments for all the necessary fonts, To create a sufficiently large set of synthetic training data, we follow the same way in [4] to render long English words sampled from a large corpus, and generate tightly cropped, gray-scale, and size-normalized text images.).
In regards to Claim 21, Cheng as modified by Wang and He further teaches the computer-implemented method of claim 20, further comprising generating the font genome data for the subset of character glyphs in the font (Wang: p. 455; more levels of representation can give rise to more abstract, more general features of the raw input, and that the lower layers of the predictor constitute a hierarchy of features that can be shared across variants of the input distribution. The authors in [7] used data from the union of all domains to learn their shared features, which is different from many previous domain adaptation methods that focus on learning features in a unsupervised way from the target domain only.).
Regarding Claim 24:
Cheng as modified by Wang and He further teaches a method comprises (Cheng: Par. (9); a method and computer-readable medium for generating resolution independent character images for display on a output device having a predefined level of resolution is disclosed.): receiving data representing an input character glyph associated with a particular font (Cheng: Par. (26); In the first step of the process in FIG. 2A, at block 302, a font designer retrieves a previously generated set of stroke-based characters.); generating first spacing data indicative of a spacing proximate to the input character glyph associated with the particular font; generating first image data comprising the data representing the input character glyph associated with the particular font and the generated first spacing data; generating second spacing data indicative of a spacing proximate to the input character glyph in the generated first image data (Cheng: Par. (44); FIG. 14A illustrates an inconsistency that may occur with four parallel strokes that must be displayed with equal spacing. The spaces between the two center vertical strokes 442 and 446 and the space between the left stroke 440 and center stroke 442 equals three pixel widths. However, the space between center stroke 446 and right stroke 448 is equal to two pixel widths. The same procedure applies as that described above for the three parallel stroke situation. However, there is just one more stroke that must be moved in case of a display error. FIG. 14B illustrates that the right stroke 448 had enough space within the character space to move and make the space between right stroke 448 and center stroke 446 equal to three, thereby producing equal spaces between the vertical strokes, Par. (28); Then, at block 314, characters are generated for display on an output device having a predefined L-level of resolution. Characters are reconstructed from the glyph construction described above. The character reconstruction process is described in FIG. 2B below.); providing, as input to a machine learning model, the generated first image data and the generated second spacing data (obvious to one skilled in the art that the generated characters from Cheng can be implemented into the classification model of Wang); obtaining, as output from the machine learning model, second image data of a set of output character glyphs associated with the particular font (Wang: p. 453; To create a sufficiently large set of synthetic training data, we follow the same way in [4] to render long English words sampled from a large corpus, and generate tightly cropped, gray-scale, and size-normalized text images.); generating a vector format of the obtained second image data of the set of output character glyphs; extracting the second spacing data from the generated vector format of the set of output character glyphs; scaling the generated vector format to match to a form of the data representing the input character glyph; and providing the scaled vector format of the set of output character glyphs for output, wherein the scaled vector format comprises the set of output character glyphs associated with the particular font (He: p. 5; As for strokes (or components), we encode each character into a 32-dimensional vector. Each dimension of the vector represents the number of corresponding basic strokes (or components) it contains (shown in Fig. 3 and Fig. 4). This count encoding can better represent the stroke (or component) attribute of a character than one-bit encoding used in StrokeGAN Zeng et al. (2021). Thereafter, a stroke (or component) vector can be expanded into a vector consistent with the dimension of the content embedding.).
Pertinent Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Jain et al. (U.S. Patent App. Pub No. 2019/0243875 A1) teaches systems and computer-implemented methods that may be used to replace missing glyphs.
Singh et al. (U.S. Patent App. Pub No. 2022/0083772 A1) teaches systems, methods, and non-transitory computer readable media for generating and providing matching fonts by utilizing a glyph-based machine learning model.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RENAE BITOR whose telephone number is (703)756-5563. The examiner can normally be reached Monday to Friday: 8:00 - 5:30 but off the 1st Friday of the biweek.
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, GREG MORSE can be reached on (571)272-3838. 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.
/RENAE A BITOR/Examiner, Art Unit 2663
/GREGORY A MORSE/Supervisory Patent Examiner, Art Unit 2698
Prosecution Insights