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 .
This action is in response to the Amendment filed on 09/02/2025.
Claims 1-2, 5-10, 13-14, 19, 21-22, 24-34 are pending. Claims 1-2, 5-6, 9-10, 14, 21, 24-26, 30, 31, 32 have been amended. Claims 3-4, 11-12, 15-18, 20, 23 have been cancelled. Claims 33-34 are newly added.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 3/5/2026 has been entered.
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.
Claim(s) 1, 19, 22, 25, 31, 32, 33, 34 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He), further in view of Cok (US 8831360 B2).
Regarding Claim 1, Shechtman teaches a method comprising: causing, by a system, presentation of a plurality of images on a user interface of the system (Shechtman, Paragraph [0036], "the term 'image' or 'digital image' refers to a digital graphics file"; Paragraph [0140], "the image composite system 1104 is located on a computing device 1100 ... the computing device 1100 may represent various types of client devices"); receiving, by the system, a selection of at least two images of the plurality of images (Shechtman, Paragraph [0089], "the image composite system receives user input that includes a background image 306 and a foreground object 308"; Paragraph [0188], "the series of acts 1600 includes an act 1610 of receiving a foreground image and a background image"), where receiving a background image and a foreground image) [[ identifying, by the system, a plurality of features of each image of the at least two images; determining, by the system, a frequency distribution of each feature of the plurality of features; and determining, based at least in part on the frequency distribution of each feature, a theme. ]] generating, by the system, an output image corresponding to the theme by blending one or more features of the plurality of features of the at least two images (Shechtman, Paragraph [0037], "the term 'composite image' refers to a digital image formed from two or more images ... a composite image includes a foreground object from a first image (real or synthetic) and a background from a second image (real or synthetic)")); causing presentation of the output image on the user interface of the system (Shechtman, Paragraph [0140], "the computing device 1100 may represent various types of client devices" together with the composite image generated as above).
But Shechtman does not explicitly disclose identifying, by the system, a plurality of features of each image of the at least two images; determining, by the system, a frequency distribution of each feature of the plurality of features; and determining, based at least in part on the frequency distribution of each feature, a theme.
determining, by the system, based at least in part on [[ the frequency distribution of ]] each feature, a theme (He, Paragraph, [0007], " the at least two frames of initial training facial sample images and the initial training facial image containing the facial information of the same person"; Paragraph [0005], " acquiring at least two frames of facial images extracted from a target video"; Paragraph [0037], "The at least two frames of facial images are facial images of a same person"; Paragraph [0038], " Step 202 includes inputting the at least two frames of facial images into a pre-trained generative model to generate a single facial image"; Paragraph [0039], "the electronic device may input the at least two frames of facial images into the pre-trained generative model to generate the single facial image").
He and Shechtman are analogous since both involve user-provided images being processed by machine-learning models to generate composite or synthesized output images. Shechtman provided a way of receiving user input and generating composite images. He provided a way of generating a single facial image from at least two input frames that contain "facial information of the same person" using a pre-trained generative model. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to incorporate multi-frame same-subject generative fusion taught by He into modified invention of Shechtman such that the device determines a theme and generates the output image using that theme. The motivation is to improve consistency and realism of the generated image by leveraging the using multiple frames containing the same person improves facial-feature fidelity.
But the combination of Shechtman, He and Cok does not explicitly disclose identifying a plurality of features of each image of the at least two images, determining a frequency distribution of each feature of the plurality of features, and [[ determining a theme ]] based at least in part on the frequency distribution of each feature.
However, Cok teaches identifying a plurality of features of each image of the at least two images (Cok, Column 7, Line 65-67-Column 8, Line 1-5, "the image types are associated with the digital images by processing and analyzing (step 555) the digital images to determine the image type(s) of the digital images (step 560), for example, by analyzing the digital images using software executing mathematical
algorithms on an electronic processor"; Column 9, Line 36-49,"An image type is a category or classification of image attributes and can be associated with a digital image as image metadata stored with the digital image in a common electronic file or associated with the digital image in a separate electronic file. An image can have more than one image type. For example, a digital image can have an image type such as a portrait orientation type, a landscape orientation type, or a scenic image type. The same digital image can also be classified as an image that includes a person type, a close-up image of a person type, a group image that includes multiple people type, day-time image type, night-time image type, image including one or more animals type, black-and-white image type, color image type, identified person type, identified gender type, and flash-exposed image type "); determining a frequency distribution of each feature of the plurality of features (Cok, Column 13, Line 40-47, "once the image types are determined for each of the digital images in the plurality of digital images, the relative frequency of digital images of each image type can optionally be determined. For example, if a collection of 60 digital images is provided and 30 are determined by the processing system to be scenic, then the relative frequency data stored in association with the collection is a value representing 50%"; Column 6, Line 46-57, "Referring to FIG. 2, a histogram of a digital image collection having a plurality of digital images of four different image types is illustrated. This kind of histogram profile is also referred to herein as an image distribution. An image distribution can be used to describe a collection of digital images in a database (collection) of images or in an image-based product, and it can be used as a filter or template to predefine a distribution of digital images, which is then used to select digital images from an image collection (or database) to be included in an image-based product. The height of each column indicates the image type count 300 of digital images of the image type marked"); determining a theme based at least in part on the frequency distribution of each feature (Cok, Column 8, Line 26-28, " the method further includes analyzing the plurality of digital images (step 570), to determine themes associated with the plurality of digital images (step 575)"; Line 43-46, " an image collection including an image of a cake with lit candles and “Happy Birthday” written on the cake is deduced to include images of a birthday party and a birthday party theme selected"; Line 48-53, " An image distribution is chosen that includes images that are available in the image collection and that correspond to an image-based product selection. In one embodiment, the relative frequency of image types in the image collection is analyzed to determine an image distribution selection "; Abstract, "providing one or more image distributions, each image distribution corresponding to a theme and including a distribution of image types related to the theme; selecting a theme having a corresponding image distribution, the image distribution having a distribution of image types").
Cok and Shechtman are analogous art since both of them are dealing with computer implemented systems that process multiple digital images to identify their visual characteristics and generate output content based on those characteristics. Shechtman provided a way of receiving user selected images and generating a composite output image by blending foreground and background features using a generative adversarial network. Cok provided a way of analyzing each digital image in a
collection to identify and classify a plurality of features (image types) per image, computing the relative frequency distribution of each feature type across those images, and determining a theme from that frequency distribution. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Co k's per-image feature identification and frequency-distribution-based theme determination into the modified invention of Shechtman such
that the system identifies a plurality of features of each of the at least two selected images, computes the frequency distribution of each feature, and determines the theme based at least in part on that frequency distribution - rather than relying solely on the implicit same-person theme of He – before generating the output image by blending features of the at least two images. The motivation is to provide a more robust, generalizable, and data-driven mechanism for theme determination that operates across diverse image content beyond facial images, leveraging.
Regarding Claim 19, the combination of Shechtman, He and Cok teaches the
invention in Claim 1.
The combination further teaches wherein the output image includes real-elements blended with generated elements (Shechtman, Paragraph [0051], “the image composite system trains the adversarial discrimination neural network 104 using real training images [0036], “images can be real, where foreground objects are naturally placed in background scenery and captured together in the same image, or synthetic (e.g., composite image) where a computing device generates one or more objects or elements of the image”).
Regarding Claim 22, the combination of Shechtman, He and Cok teaches the
invention in Claim 1.
The combination further teaches wherein the at least two images include at least one of real images or generated images (Shechtman, Paragraph (0122], "the foreground objects and/or background images can be synthetically generated or acquired from real images") the real-images are defined using real-elements captured from real world environment (Shechtman, Paragraph (0036], "images can be real, where foreground objects are naturally placed in background scenery and captured together in the same image") and the generated images are virtual elements defined using generated elements defined to mimic a look and behavior of corresponding real-elements." (Shechtman, Paragraph [0031], "generates composite images that include geometrically correct spatial transformations of foreground objects without the need of user manipulation. Indeed, the image composite system generates natural and realistic looking composite images").
Regarding Claim 25, the combination of Shechtman, He and Cok teaches the
invention in Claim 1.
The combination further teaches blending the one or more features includes mixing and arranging the one or more features (Shechtman, Paragraph [0037], "a composite <read on blended> Image includes a foreground object from a first <read on feature> image ... and a background from a second <read on feature> image (real or synthetic)."; Paragraph [0024], "a geometric prediction neural network using an adversarial discrimination neural network to learn warp parameters that provide correct geometric alignment of foreground objects with respect to a background image") of the at least two images (Schechtman, Paragraph [0040], “machine learning may operate by building models from example inputs (e.g., training), such as a training image set”).
Regarding Claim 31, it recites limitations similar in scope to the limitations of claim 1 and the combination of Shechtman, He and Cok teaches all the limitations as of Claim 1. And Shechtman discloses these features can be implemented on a computer readable storage medium (Shechtman, Fig. 11, Paragraph [0151], “the components 1106-1122 can include one or more instructions stored on a computer-readable storage medium and executable by processors of one or more computing devices, such as a client device or server device. When executed by the one or more processors, the computer-executable instructions of the image composite system 1104 can cause the computing device(s) to perform the feature learning methods described herein”).
Regarding Claim 32, it recites limitations similar in scope to the limitations of claim 1, but in a system. As shown in the rejection, the combination of Shechtman, He and Cok disclose the limitations of claims 1. Additionally, Shechtman discloses an system that maps to Fig. 11 and Paragraph [0143], [0151] (Shechtman, Fig. 11, Paragraph [0143], “As illustrated in FIG. 11, the image composite system 1104 includes various components. For example, the image composite system 1104 includes an image and object manager 1106, a generative adversarial network 1108, an image compositor 1114, and a storage manager 1116; [0151], “the components 1106-1122 can include one or more instructions stored on a computer-readable storage medium and executable by processors of one or more computing devices, such as a client device or server device”) Thus, Claim 11 is met by Shechtman according to the mapping presented in the rejection of claims 1, given the method corresponds to the system.
Regarding Claim 33, the combination of Shechtman, He, and Cok teaches the invention in Claim 1.
The combination further teaches determining a plurality of weights associated with the plurality of features, wherein the frequency distribution of each feature is based at least in part on the plurality of weights; each weight of the plurality of weights corresponds to a feature of the plurality of features; and generating the output image is based at least in part on the plurality of weights (Cok, Column 9, Line 56-64 " An image type can include a value that indicates the strength or amount of a particular type for a specific image. For example, an image is a group image but, if it only includes two people, the strength of the group-type is relatively weak compared to a group image that includes 10 people. In this example, an integer value representing a number of persons appearing in the digital image is stored with or in association with the digital image to indicate its group-type strength or value"; Column 10, Line 4-9, "An image-usage type can have a strength value indicating how often or how much the corresponding digital image is used, for example including a combination of metrics such as how often the image is shared or viewed, whether the image was purchased, edited, used in products, or whether it was deleted from a collection"; Column 11, Line 40-43, "once the image types are determined for each of the digital images in the plurality of digital images, the relative frequency of digital images of each image type can optionally be determined"; Column 15, Line 53-56, "using a computer to select digital images from the plurality of digital images, the selected digital images having the image distribution corresponding to the selected theme").
Cok and Shechtman are analogous for the same reasons set forth in the rejection of Claim 1 above.
Regarding Claim 34, the combination of Shechtman, He, and Cok teaches the invention in Claim 33.
The combination further teaches wherein the output image is a first output image generated based at least in part on a first weight of the plurality of weights (Cok, Column 10, Line 32-38, “a first desired image distribution specification can include 20% scenic images, 60% scenic images that include a person, and 20% close-up images. The actual number of images of each type is then calculated by multiplying the total number of images in the desired image-based product by the percentage associated with the image type in the desired image distribution”); and the method further comprises: generating, by the system, the first output image includes providing one or more weight adjusters that can be interacted with to adjust generation of subsequent output images, wherein each weight adjuster of the one or more weight adjuster corresponds to a corresponding weight of the plurality of weights (Cok, Column 15, Line 22-27, " A processor can be used to provide a user interface, the user interface including controls for setting the relative frequencies of digital images of each image type"; Line 24-27, " a preferred method of the present invention can include using a processor to receive a distribution of image types that includes a range of relative frequencies of image types"); receiving, by the system, one or more user inputs interacting with a first weight adjuster of the one or more weight adjusters to adjust the first weight that is associated with the first weight adjuster to a second weight of the plurality of weights; (Cok, Column 15, Line 24-27, "a preferred method of the present invention can include using a processor to receive a distribution of image types that includes a range of relative frequencies of image types"); generating, by the system, based at least in part the second weight, a second output image corresponding to the theme (Cok, Column 5, Line 60-61, “assembling the selected digital images into an image based product in step 525”; Column 9, Line 2-9, “A large image collection can include images of many different, unrelated events. Different subsets of the digital images in the image collection can correspond to different themes… A single theme can have different image distributions”); causing, by the system, presentation of the second output image on the user interface of the system (Cok, Column 5, Line 45-48, “Such display can be interactively controlled by a user. Such display devices and image-based products are known in the art as are user interfaces for controlling the viewing of image-based products on a display”)
Cok and Shechtman are analogous for the reasons stated in the rejection of Claim 1 above.
Claim(s) 2, 5, 6, 8, 9, 10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He) further in view of Cok (US 8831360 B2) as applied to Claim 1 above and further in view of Zhang et al. (US 20160335789 A1, hereinafter Zhang).
Regarding Claim 2, the combination of Shechtman, He and Cok teaches the invention in Claim 1.
The combination further teaches prior to presenting the output image (Shechtman, Paragraph [0037], "a composite <read on output> image includes a foreground object from a first <read on image> ... and a background from a second <read on image>"; Paragraph [0140], "I/O interfaces ... provide graphical data to a display for presentation to a user.") receiving one or more additional inputs (Shechtman, Paragraph [0089], "the image composite system receives user input that includes a background <read on image> 306 and a foreground object 308")
But Shechtman does not explicitly disclose additional inputs to adjust at
least one feature of the output image ... based at least in part on the theme,
generating a second output image by adjusting the at least one feature of the
output image ... presenting the second output image in lieu of the output image
However, He teaches generation occurring prior to presenting the output image (He, Paragraph [0038], "Step 202 includes inputting the at least two frames of facial <read on images> ... to generate a single facial image."; Paragraph (0032], "The backend server may return data (such as image data) to the terminal devices, for presentation on the terminal devices.") based at least in part on the theme, generating ... an output image (He, Paragraph [0036], "Step 201 includes acquiring at least two frames of facial <read on images> extracted from a target video." Paragraph [0037], "The at least two frames of facial <read on images> are facial images of a same person." Paragraph (0038], "inputting the at least two frames ... into a pre-trained generative model to generate a single facial image.")
He and Shechtman are analogous because both involve computer implemented
imagery systems that receive image inputs, apply processing (compositing or generative modeling), and present an image to a user. Shechtman provided a way of receiving user input specifying images and generating a composite displayed to a user
. He provided a way of using multiple images of a same subject (theme) and generating a single output image based on that theme, then presenting it. Therefore, it would have been obvious before the effective filing date to incorporate He's theme-based generative output image into Shechtman's composite-image framework so that initial output images are generated "based at least in part on the theme," with the motivation of improving realism and consistency in composite imagery, as discussed by He (theme via same-person frames) in Paragraph [0036]-[0037].
But the combination does not explicitly disclose additional inputs to adjust
at least one feature of the output image .. second ... image ... by adjusting the at
least one feature of the output image presenting the second output image in lieu
of the output image However, Zhang teaches receiving additional inputs to adjust at least one feature of the output image (Zhang, Paragraph [0017], "receiving first user
input. .. performing a first image displaying a first image editing operation
on the first image to generate a second image ... displaying the second
image ... receiving second user input... performing a second image editing
operation on the second image to generate a third image based on the
second user input."; The second user input <read on "additional inputs">
and the second image editing operation <read on "adjust at least one feature
of the output image.">) additional inputs to adjust at least one feature of the output
image ... based at least in part on the theme, generating a second output image (Zhang, Paragraph [0005], " ... the first image layer and the second image layer may be independently edited ... The image editing operation may include changing a color attribute ... As another example ... blurring the first image layer" [0014), "performing a second image editing operation on the second image to generate a third image <read on "second output image"> based on the second user input."; The third image <read on "second output image") presenting the second output image in lieu of the output image (Zhang, Paragraph [0014), "displaying the second image at the mobile device") and by implication from the image-editing operation flow, the third image replaces the second as the newly displayed image <read on presenting the second output image in lieu of the output image>”).
Zhang and Shechtman are analogous because both involve image processing systems in which user input controls the generation or modification of images displayed on a device. Shechtman provided a way of generating composite images from multiple inputs and presenting them to the user. Zhang provided a way of taking an already-generated image, receiving additional user input, adjusting specific visual features (attributes) of that image, generating a new image version, and displaying that new version in place of the old. Therefore, it would have been obvious to incorporate Zhang's iterative feature-adjustment editing operations into Shechtman's composite image system-after the initial output image is generated-so that the system receives additional inputs, adjusts features, generates a second output image reflecting those adjustments, and presents the new image in lieu of the original. The motivation is to allow user-driven refinement of generated images prior to presentation, improving user control and visual quality, as taught by Zhang's independent editing of image attributes in Paragraph [0005].
Regarding Claim 5, the combination of Shechtman, He, Cok and Zhang teaches the invention in Claim 2.
The combination further teaches [[ prior to receiving the one or more additional
inputs]], analyzing the output image to identify a plurality of portions wherein each portion of the plurality of portions corresponds to at least one [[ feature of the plurality of features of the at least two image ]] wherein the one or more [[ additional inputs ]] include inputs to [[ at least one portion of the plurality of portions ]] (Shechtman, Paragraph [0037], "a composite <read on output> image includes a foreground object from a first <read on image> ... and a background from a second <read on image> (real or synthetic)"; it is noted composite image naturally separates foreground object and background image).
But Shechtman does not explicitly disclose prior to receiving the one or more
additional inputs .. .feature of the plurality of features of the at least two image ... additional inputs ... adjust at least one portion of the plurality of portions.
However, Zhang teaches prior to receiving the one or more additional inputs, analyzing the output image to identify a plurality of portions wherein each portion of the plurality of portions corresponds to at least one feature of the plurality of features of the at least two image wherein the one or more additional inputs include inputs to adjust at least one portion of the plurality of portions (Zhang, Paragraph [0005], "The first image layer and the second image layer <read on plurality of portions> may be independently edited by a user to create one or more visual effects. To illustrate, a user may perform an image editing operation on the first image layer but not the second image layer (or vice versa) . ... The image editing operation may include changing a color attribute of the first image layer but not the second image layer ... As another example, the image editing operation <read on one or more
additional inputs > may include blurring the first image layer but not the second image layer", Paragraph [0014], a method includes displaying a first image at a mobile device. The method further includes receiving first user input at the mobile device . ... Based on the first user input, a first image editing operation is performed on the first image to generate a second image. The method further includes displaying the second image at the mobile device and receiving second user input at the mobile device ... The method further includes performing a second image editing operation on the second image to generate a third image based on the second user input")
Zhang and Shechtman are analogous since both disclose image-processing
systems where the user sees an image and parts of it can be modified. Shechtman provides the theme-based composite output image from multiple images. Zhang provides the explicit separation into multiple layers (portions) and portion-specific editing based on user inputs. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to incorporate portion-based editing so that, prior to receiving the additional inputs taught by Zhang into modified invention of Shechtman such that the system identifies portions corresponding to features of the original images and the additional inputs adjust at least one of those portions.
Regarding Claim 6, the combination of Shechtman, He, Cok and Zhang teaches the invention in Claim 5.
The combination further teaches each portion of the plurality of portions
(Shechtman, Paragraph [0037], "a composite image includes a foreground object
from a first ... and a background from a second).
Shechtman does not explicitly disclose but Zhang teaches each portion of the
plurality of portions includes a plurality of second features that specific to each
portion; and the one or more additional inputs include inputs to adjust at least
one second feature of the plurality of second features (Zhang, Paragraph [0005], "The first image layer and the second image <read on each portion of the plurality of portions> layer may be independent[ly] edited by a user to create one or more visual effects ... The image editing operation may include changing a color attribute of the first image layer but not the second image layer ... As another example, the image editing operation may include blurring the first image layer but not the second image layer"; Paragraph [0014]/[0017], "displaying a first image at a mobile device. The method further includes receiving first user input ...
Based on the first user input, a first image editing operation is performed
on the first image to generate a second image . ... displaying the second
image ... receiving second user input ... performing a second image editing
operation on the second image to generate a third image based on the
second user input.")
Zhang and Shechtman are analogous since both of them are dealing with
image processing systems in which user input controls properties of images or
portions of images displayed on a device. Shechtman provided a way of generating a composite image that includes different portions (foreground/background) from different images and presenting that image to a user. Zhang provided a way of representing an image as portions (image layers) that each include multiple attributes/features (e.g. , color, blur) and of adjusting those attributes using additional user inputs to generate new images. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Zhang's per-layer, per-feature editing into the composite image produced by Shechtman such that each portion of the plurality of portions includes a plurality of second features that specific to
each portion, and the one or more additional inputs include inputs to adjust at
least one second feature of the plurality of second features, as recited in Claim 6.
The motivation is to allow finer-grained user control over specific features within
each portion of the composite image, improving flexibility and image quality, as
suggested by Zhang's discussion of independently editing attributes to create
desired visual effects (Paragraph [0005)).
Regarding Claim 8, the combination of Shechtman, He, Cok and Zhang teaches the invention in Claim 2.
The combination further teaches wherein receiving the one or more the additional
inputs include inputs of at least one of [[ text inputs, ] image inputs, graphic inputs, [[ audio inputs, or ]] selection inputs provided to a generative Al model (Shechtman, Paragraph [0089), "the image composite system receives user input that includes a background <read on image> 306 and a foreground object 308"; Paragraph [0090], "Using the learned weights and trained algorithms, the trained geometric prediction neural network 302 determines warp parameters for the foreground object 308 ... based on the geometric perspective of the background <read on image> 306."; Paragraph [0028], "an input image is a real image or a fake image" [0036], "the term "image" or "digital image" refers to a digital graphics
file").
Regarding Claim 9, the combination of Shechtman. He, Cok and Zhang teaches the invention in Claim 2.
The combination further teaches wherein receiving selection of the at least two
images includes receiving a selection of a portion of each image of the at least two images (Shechtman, Paragraph (0037], "a composite <read on output> image includes a foreground object from a first <read on image> ... and a background from a second <read on image> (real or synthetic)"; Paragraph (0089], "the image composite system receives user input that includes a background <read on image> 306 and a foreground object 308.").
But Shechtman does not explicitly disclose corresponding to at least one feature of the plurality of features; and generating the output image includes blending the one or more features of the plurality of features is based at least in part on the at least one feature of the selection of the portion of each image.
However, Zhang teaches receiving selection of the at least two images includes receiving a selection of a portion of each image of the at least two images corresponding to at least one feature of the plurality of features (Zhang, Paragraph (0005], "The first image layer and the second image layer may be independent[ly] edited by a user to create one or more visual effects ... The image editing operation may include changing a color attribute of the first image layer but not the second image layer ... As another example, the image editing operation may include blurring the first image layer but not the second image layer) and generating the output image includes blending the one or more features of the plurality of features is based at least in part on the at least one feature of the selection of the portion of each image (Zhang, Paragraph [0014]/(0017] reinforces that additional inputs are used to perform editing operations on the selected image (and its portions), generating new images based on those feature adjustments.
Zhang and Shechtman are analogous because both involve image processing frameworks where parts of images (foreground/background or layers) are treated as separable portions subject to user control. Shechtman provided a way of selecting which foreground and background images form the portions of the composite image and presenting that composite. Zhang provided a way of receiving user selection of which image portion/layer to edit and then using user inputs to adjust attributes (features) of that selected portion. Therefore, it would have been obvious to one of ordinary skill in the art to combine Zhang's selection-and-per-portion feature editing with Shechtman's composite image such that receiving the one or more additional inputs comprises receiving selection of a portion of each image, and the one or more additional inputs are used to adjust at least one feature of the selection of the portion of each image, as recited in Claim 9. The motivation is to give users direct control over which part of the composite is edited and which specific feature of that selected part is adjusted , improving flexibility and user satisfaction, consistent with Zhang's goals of independently editing image layers to create desired visual effects.
Regarding Claim 10, the combination of Shechtman, He, Cok and Zhang teaches the invention in Claim 9.
The combination further teaches wherein the one or more additional inputs include selecting a portion of the output image; and generating the second output image includes, on the portion of the output image, at least one of adding a new feature, adjusting an existing feature of the plurality of features, replacing the existing feature with the new feature, or excluding the existing feature (Zhang, Paragraph [0005], "The image editing operation may include changing a color attribute of the first image layer but not the second image layer" Paragraph [0005], "As another example, the image editing operation may include blurring the first image layer but not the second image layer")
Zhang and Shechtman are analogous since they both describe image systems controlled by user inputs that determine how the image appears. Shechtman provided a way of generating and presenting composite images from multiple source images, with user input selecting those sources. Zhang provided a way of modifying specific attributes/features (such as color and blur) of selected image portions/layers, including changing attributes and choosing not to apply those attributes to other layers. Therefore, it would have been obvious to one of ordinary skill in the art to apply Zhang's attribute-replacement and attribute-exclusion behavior to the portions selected and adjusted in the composite image of Shechtman, such that the one or more additional inputs include inputs to replace the existing feature with the new feature, or excluding the existing feature. The motivation is to allow users to not only adjust features of selected portions, but to explicitly replace or omit those features as needed to achieve the desired visual style, which is consistent with Zhang's teaching of independently editing attributes to create multiple visual effects.
Claim(s) 7, 26, 27 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He) further in view of Cok (US 8831360 B2) as applied to Claim 1 above and further in view of Systrom et al. (US 20140279068 A1, hereinafter Systrom).
Regarding Claim 7, the combination of Shechtman, He and Cok teaches the base
method of Claim 1.
The combination does not explicitly disclose but Systrom teaches the plurality of
images presented at the user interface are uploaded by a user or are retrieved from one
or more content sources (Systrom, Paragraph [0023], " the image can be an "official" image, such as uploaded by a brand, merchant, store, or entity thereof as part of an advertising campaign, or the image can be an "unofficial" image, such as uploaded by a private user and tagged with various metadata by the private user and/or any other private user. In one implementation, Block S110 uploads the image from a standalone computing device, such as a desktop computer, a tablet, a smartphone, an Internet-capable camera, etc. In another implementation, Block S110 uploads the image from a local or remote server. For example, Block S110 can retrieve the image from any other Internet site by accessing the image through a server that supports the other Internet site"; Paragraph [0136], "A user can send a request to the web server 832 to upload information, for example, images or videos that are stored in the content store 808.") the one or more content sources include a social media account of the user accessed over the social media account or a website of a content provider accessed over a network (Systrom, Paragraph [0025], "a first user can capture a photographic image using the camera integrated into his mobile device, and Block 5110 can upload the photographic image to a social networking system and post the photographic image to a social feed within the social networking system."; Paragraph [0136], "The web server 832 links the social networking system 704 via the network 740 to the client device 708, ... A user can send a request to the web server 832 to upload information, for example, images or videos that are stored in the content store 808.") the images in the social media account are generated by the user or shared by the user or other users, the plurality of images retrieved and uploaded from the social media account in accordance to privacy setting defined by the user (Systrom, Paragraph [0023], "the image can be an 'official image,' such as uploaded by a brand, merchant, store, or entity that designs, manufactures, produces, sells, leases, promotes, or otherwise distributes content associated with the image. Alternatively, the image can be an 'unofficial image,' such as uploaded by a private user and/or any other private user within (or outside) the social networking system"; Paragraph [0137], "The image can be one form of user generated 'content accessible through the social networking system to enhance a user experience. …Thus a user of the social networking system may be encouraged to communicate with another user by 'posting content items of one or more types through various communication channels within the social networking system ... a 'stream' through which a series of content items posted, uploaded, or otherwise provided to the social networking system are shared with one or more users." ; Paragraph [0138], "Content sharing between users can also be limited. For example, a user can post a video from a company presentation to the Social networking system, wherein the video is not appropriate for sharing across all of the user's connections (e.g., connections that include employees of competing companies) the images are identified and retrieved from the website using links provided by the user at the user interface (Systrom, Paragraph [0053], "Block 5160A ... functions to implement a hotspot in the image by opening or linking to additional brand or product-related content. For example, when a user clicks, touches, or otherwise selects the brand-tagged image or a hotspot within the image, Block S160A can direct the user to another page, menu, or interface within the Social networking system ... Alternatively, when a user clicks, touches, or otherwise selects the brand-tagged image or a hotspot within the image, Block S160A can direct the user outside of the Social networking system, such as to a brand website, a blog associated with the brand, a brand-related native application, or an online standalone store for the brand."; Paragraph [0023), "the system can retrieve the image from any other Internet site by accessing the image through a server that supports the other Internet site").
Systrom and Shechtman are analogous since both of them are dealing with
networked image-based user interfaces where users interact with images delivered over
a network, and the systems manage image content stored on servers and presented on
client devices. Shechtman provided a way of generating composite images using a
generative adversarial network and presenting the resulting images on client devices via
a web-hosted image composite system. Systrom provided a way of loading images to a
social networking system, having users (or brands) upload images, posting them to
social feeds, limiting sharing, and linking from those images to brand content or
websites. Therefore, it would have been obvious to one of ordinary skill in the art before
the effective filing date of the claimed invention to incorporate Systrom's teachings
about using social networking accounts and external websites as content sources,
obtaining user-generated I user-shared images from those sources subject to user defined sharing limitations, and using user-activated links (hotspots) to retrieve content from external websites into the Shechtman image composite system such that the
plurality of images presented at the user interface are uploaded by a user or retrieved
from social media accounts or content-provider websites, in accordance with user
privacy preferences, and identified and retrieved via links provided by the user at the
interface. The rotivation is to leverage existing social-media content and user
interactions to supply images and branded content into an image-processing experience
and to respect users' desired sharing scope-improving personalization and
commercial relevance of the composite-image system-as discussed by Systrom in
connection with user-uploaded images, content items, limited sharing, and link-based
navigation to brand content (e.g ., Paragraphs [0023], [0050]-[0053], [0136]-[0138]).
Regarding Claim 26, the combination of Shechtman, He and Cok teaches the
invention in Claim 1.
The combination further teaches generating the output image includes generating the output image [[ based at least in part on historical user data of a user]] of the system (Shechtman, Paragraph [0037], "a composite <read on output> image includes a foreground object from a first <read on image> ... and a background from a second <read on image>." [0153], "the environment 1200 includes various computing devices including server device(s) 1202 and one or more client devices 1204a, 1204b").
But the combination does not explicitly disclose based at least in part on historical user data of a user [[ of the client device ]].
However, Systrom teaches generating the output image includes generating the output image based at least in part on historical user data of a user of the system (Systrom, Paragraph [0125], "Each user of the social networking system 704 is associated with a user profile, which is stored in the user profile store 804. A user profile includes declarative information about the user ... A user profile can also store other information provided by the user, for example, images or videos."; Paragraph (0127], "The content store 808 stores content items associated with a user profile, such as images, videos or audio files. Content items from the content store 808 can be displayed when a user profile is viewed"; Paragraph [0130)-(0131), "The action log 816 can be used ... to track user actions <read on historical user data of a user> ... Information describing these actions can be stored in the action log 816 .. . data from the action log 816 is used to infer interests or preferences of the user, augmenting the interests included in the user profile and allowing a more complete understanding of user preferences.").
Systrom and Shechtman are analogous since all involve computer-implemented
systems that generate or present visual content to users based on stored user-specific
information. Shechtman provided ways of generating an output image from multiple input images and presenting it to a user. Systrom provided a way of maintaining user profiles, content items (images, videos, audio files) and action logs that capture user actions and inferred preferences, which constitute historical user data used to tailor what content is presented. Therefore, it would have been obvious before the effective filing date to incorporate user-profile and history mechanisms taught by Systrom into modified invention of Shechtman such that generating the output image includes generating the output image based at least in part on historical user data of a user of the client device, with the motivation of personalizing generated images to match the user's past media and inferred preferences, as described in Systrom's discussion of user profiles, content store, and action log.
Regarding Claim 27, the combination of Shechtman, He, Cok and Systrom teaches the invention in Claim 26.
The combination further teaches the wherein the historical user data includes
past media content associated with the user (Systrom, Paragraph [0125), "A user profile can also store other information provided by the user, for example, images or videos. Images of users can be tagged with identification information of users of the social networking system 704 displayed in an image."; Paragraph [0127], "The content store 808 stores content items associated with a user profile. Such as images, videos or audio files. Content items from the content store 808 can be displayed when a user profile is viewed or when other content associated with the user profile is viewed.").
As explaimed in rejection of Claim 26, the obviousness for combining of historical data of Systrom into Shechtman is provided above.
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He) further in view of Cok (US 8831360 B2) as applied to Claim 1 above and further in view of Zhang et al. (US 20160335789 A1, hereinafter Zhang) and Systrom et al. (US 20140279068 A1, hereinafter Systrom).
Regarding Claim 13, the combination of Shechtman, He, Cok and Cok teaches the invention in Claim 1.
The combination further teaches selecting a region of the output image for the
theme (Shechtman, Paragraph [0089], "the image composite system receives user
input that includes a background image 306 and a foreground object 308";
Paragraph [0037], "a composite image includes a foreground object from a first
image ... and a background from a second image")
But Shechtman does not explicitly disclose theme to generate a decal, wherein
the decal generated is associated with a user profile of a user.
However, Zhang teaches selecting a region of the output image for the theme to
generate a decal (Zhang, Paragraph [0116], "The method 1200 includes receiving
first user input from a user interface ... The first user input indicates an image for
a display operation."; Paragraph [0118], "The method 1200 may further include
receiving second user input from the user interface ... The second user input
identifies a first image layer of the image ... The second user input may identify a
foreground of the image using a swipe action at a touchscreen device.";
Paragraph [0121], "The method 1200 may further include receiving third user
input from the user interface ... The third user input identifies a second image
layer of the image ... The background may correspond to the second image layer"; it is noted the edited layer used in subsequent display is decal generated from that region).
Zhang and Shechtman are analogous since both deal with computer implemented image systems in which user input selects portions of images (foreground, background, layers) and the system generates edited/composite images for display. Shechtman provided a way of generating composite images by taking user-selected foreground/background content and using a generative network to output a composite image. Zhang provided a way of receiving user input that selects specific image layers and performing editing operations on those layers. Therefore, it would have been obvious before the effective filing date to incorporate Zhang's layer-selection and editing operations into the modified invention of Shechtman such that, after the output image for the theme is generated, the system enables selecting a region of the output image for the theme to generate a decal (a selected sub-image or layer used as an overlay). The motivation is to give users finer control over localized edits and overlays on composite images, improving flexibility and visual customization, as discussed in Zhang's independent layer selection and editing in Paragraphs [0116], [0118] and [0121].
The combination does not explicitly disclose but Systrom teaches wherein the decal generated is associated with a user profile of a user (Systrom, Paragraph [0125], "Each user of the social networking system 704 is associated with a user profile, which is stored in the user profile store 804."; Paragraph (0122)-(0123], "A user profile can also store other information provided by the user, for example, images or videos. Images of users can be tagged with identification information of <read on decal generation> users of the social networking system 704 displayed in an image.")
Systrom and Shechtman are analogous because both concern systems where
user-provided images (or edited regions of images) are stored and managed in
association with user accounts/profiles. Shechtman provide the mechanism to generate
an output image and select a region of that image (a decal). Systrom provides the
mechanism to store images and associate them with user profiles, including tagging
images with user identification information. Therefore, it would have been obvious to
one of ordinary skill in the art before the effective filing date to associate the decal
generated from the selected region of the output image with a user profile of a user, as
in Systrom's user profile store, so that the decal is stored and used as content
associated with the user's profile. The motivation is to enable the decal to function as
user-specific content (e.g., profile imagery or visual identifier) within a social or
interactive system, leveraging known user-profile/image associations in Systrom.
Claim(s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He), further in view of Cok (US 8831360 B2), Zhang et al. (US 20160335789 A1, hereinafter Zhang) and Systrom et al. (US 20140279068 A1, hereinafter Systrom) as applied to Claim 13 above and further in view of Ghosh et al. (US 20100281427 A1, hereinafter Ghosh) and Blattner (US 20110148916 A1).
Regarding Claim 14, the combination of Shechtman, He, Cok, Zhang, and Systrom teaches Claim 13,
The combination further teaches identifying, based at least in part on the decal,
the user (Systrom, Paragraph [0125], "Each user of the social networking system
704 is associated with a user profile, which is stored in the user profile store
804."; Paragraph [0122)-(0123], "A user profile can also store other information
provided by the user, for example, images or videos. Images of users can be
tagged with identification information of users of the social networking system
704 displayed in an image.").
As explained in rejection of Claim 13, the obviousness for combining of the decal
generated from the selected region of the output image with a user profile of a user of
Systrom into Shechtman is provided above.
But this combination does not explicitly disclose but Ghosh teaches to
automatically load settings and preferences of the user at an interactive application
(Ghosh, Paragraph [0030], "Once a user profile persona is selected, the user
profile persona can be communicated to the touch point 104 at which the user is
accessing the service."; Paragraph [0034], "Next, the selected user profile
persona is provided (at 208) by the device 100 to the touchpoint 104 to enable
customization of a service being accessed by the user.")
Ghosh and Shechtman/Zhang/Systrom are analogous since all involve user specific data being used to personalize how services or content are presented.
Shechtman/Zhang/Systrom provide a way to generate an image decal from the output
image and associate that decal with a user profile, thus allowing the user to be identified
from the decal. Ghosh provides a way to use a selected profile persona to customize a
service automatically at a touchpoint. Therefore, it would have been obvious before the
effective filing date to use the association between the decal and the user profile (as in
Systrom) to identify the user and then, following Ghosh, load the user's settings and
preferences (profile persona) automatically into the interactive application, so that the
method performs "identifying, based at least in part on the decal, the user and to
automatically load settings and preferences of the user at an interactive application."
The motivation is to improve user experience by providing automatic profile-based
customization once the user is recognized visually by the decal, consistent with Ghosh's
profile-persona-based service customization.
The combination still does not explicitly disclose but Blattner teaches identifying,
based at least on the decal, a miniature theme and is used as a background for an
avatar representing the user in the interactive application (Blattner, Paragraph [0107],
"The personality section 1002 of the avatar settings interface 1000 includes an
avatar list 1015 including the one or more various avatars corresponding to the
user of the instant messaging system."; Paragraph [0109], "The behavior of the
avatar is summarized in a card front 1045 and a card back 1050 displayed on the
personality section. The card front 1045 includes an illustration of the avatar and
wallpaper over which the avatar 1020 is illustrated.")
Blattner and Shechtman are analogous because both involve user-specific visual representations (avatars or image-based symbols) that can be customized in appearance and used to represent the user in interactive communication environments. Shechtman provide the decal derived from the theme based output image, associated with the user profile. Blattner provides a way of presenting an avatar representing the user together with a background "wallpaper" on which the avatar is illustrated. Therefore, it would have been obvious before the effective filing date to use the decal (a region of the themed output image) as a miniature theme and apply it as the background wallpaper behind the user's avatar in the interactive application taught by Blattner into modified invention of Shechtman such that the method performs "identifying, based at least on the decal, a miniature theme and is used as a background for an avatar representing
the user in the interactive application." The motivation is to visually integrate the theme based decal into the user's avatar representation, providing consistent visual branding
or personalization across the user's profile and avatar, as suggested by Blattner's use
of wallpaper and avatar customization.
Claim(s) 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He) further in view of Cok (US 8831360 B2) as applied to Claim 1 above and further in view of Zhang et al. (US 20160335789 A1, hereinafter Zhang) and Ghosh et al. (US 20100281427 A1, hereinafter Ghosh).
Regarding Claim 21, the combination of Shechtman, He and Cok teaches the invention in Claim 1
The combination does not explicitly disclose but Zhang teaches the output image
comprises a plurality of versions, each of the plurality of versions generated based at
least in part on the theme and an identity of a corresponding other user (Zhang, Paragraph [0017], "receiving first user input... performing a first image editing operation on the first image to generate a second image ... displaying the second image ... receiving second user input... performing a second image editing operation on the second image to generate a third image based on the second user input"; it is noted the first image, second image and third image are a plurality of vr sions of the output image").
Zhang and Shechtman are analogous since both deal with computer implemented image editing systems in which a user selects portions of an image and modifies those portions, and the modified images are presented on a device. Shechtman provided a way of generating composite images by taking user-selected foreground/background content and using a generative adversarial network to output a composite image. Zhang provided a way of receiving user input that selects specific image version of image and performing editing operations on those versions. Therefore, it would have been obvious before the effective filing date to incorporate Zhang's version of selection and editing operations into modified invention of Shechtman such that after the output image for the theme is generated, the system enables selecting a different version of selection and process the image accordingly. The motivation is to give users finer control over personalized edits and overlays on composite images, improving flexibility and visual customization, as discussed by Zhang's independent adjustment of foreground/background attributes in Paragraphs [0049]- [0053].
But the combination of does not explicitly disclose a particular version of the plurality of versions of the output image is presented in accordance with a determined context of an interactive application;
However, Ghosh teaches a particular version of the plurality of versions of the
output image is presented in accordance with a determined context of an interactive
application (Ghosh, Paragraph [001 0], "multiple user profile personae associated
with the user are maintained such that different ones of the user profile personae
are selected by a profile selection mechanism based on the context of accessing
the particular service"; [0015], "the context in which a user accesses a service
impacts the user's behavior, preferences, and/or interests"; [0033]-[0034], "the
device 100 selects (at 206) one of the plural user profile personae to use for the
service, based on the context ... the selected user profile persona is provided (at
208) ... to enable customization of a service being accessed by the user")
Ghosh is analogous to Shechtman because all describe systems in which multiple representations associated with a user (images or profiles) are maintained and one is selected for presentation based on circumstances. Shechtman provided a way of
generating an output composite image from multiple images for a user and displaying
that image. Ghosh provided a way of maintaining multiple user profile personae and
selecting one persona based on context to customize the service presented to the user.
Therefore, it would have been obvious before the effective filing date to incorporate
Ghosh's context-based persona selection into the image-generation framework of
Shechtman such that multiple versions of the output image (each corresponding to a
different user qr user identity) are generated based on a theme and user identity, and a
particular version is selected and presented according to the determined context of the
interactive application, with the motivation of providing context-appropriate visual
representations tailored to different users, as taught by Ghosh's context-driven persona
selection.
Claim(s) 24, 28-30 is/are rejected under 35 U.S.C. 103 as being unpatentable over Shechtman et al. (US 20190251401 A1, hereinafter Shechtman), in view of He et al. (US 20190080148 A1, hereinafter He), further in view of Cok (US 8831360 B2) as applied to Claim 1 above and further in view of Ghosh et al. (US 20100281427 A1, hereinafter Ghosh).
Regarding Claim 24, the combination of Shechtman, He and Cok teaches the
invention in Claim 1.
The combination further teaches wherein the output image is dynamically updated (Shechtman, Paragraph [0140), "1/0 interfaces 1420 may provide devices such as a graphical display 1422 ... to provide graphical data to a display for presentation to a user."; [0067], "Accordingly, the image composite system employs iterative updates of the warp parameters, which makes smaller incremental improvements as well as consistently yields better results").
But combination does not explicitly disclose the output image is dynamically
updated [[ based on changes to context of a interactive application]].
However, Ghosh teaches wherein the output image is dynamically updated
based on changes to context of a interactive application (Ghosh, Paragraph (001 0],
"multiple user profile personae associated with the user are maintained such that
different ones of the user profile personae are selected by a profile selection
mechanism based on the context of accessing the particular service." Paragraph
(0011], "if the user is accessing the service during the weekend, a first of the user
profile personae may be selected ... when the user is accessing the service
during a week day, another one of the user profile personae can be selected for
accessing the service."; Paragraph (0015], "the con text in which a user accesses
a service impacts the user's behavior, preferences, and/or interests when
accessing the service. The context includes the environment of the user (e.g.,
location of the user), time information, tasks being performed by the user ... a
social network of the user, and so forth."
Ghosh rnd Shechtman are analogous since both deal with computer implemented systems that present information to a user and tailor that information based on user-related parameters. Shechtman provided a way of presenting an output image on a client device UI. Ghosh provided a way of determining a context and selecting among multiple user profile personae based on that context so that the information used during access of the service changes with the context. Therefore, it would have been obvious before the effective filing date to incorporate Ghosh's contextbased persona selection into Shechtman's image-presentation framework such that the output image is dynamically updated based on changes to context of a interactive application, with the motivation of tailoring what is shown to the user to suit their current environment, time, task, and social situation, as discussed by Ghosh in Paragraph [0015].
Regarding Claim 28, the combination of Shechtman, He and Cok teaches the invention in Claim 1.
The combination does not explicitly disclose but Ghosh teaches determining an
environmental context of the output image based at least in part on user interaction in a virtual environment; and updating, based on the environmental context, the output image to generate a second output image (Ghosh, Paragraph [0030] ... context engine 106 determines a current context for a u er based on signals representing the user's interactions with one or more services ... " "[0032] ... services 120 may include communication services, social networking services, online games, or other network-based applications accessed via user devices 102.") where the "services ... online games ... social networking services ... " <read on a virtual environment>, and "determines a current context ... based on ... interactions" <read on determining an environmental context ... based at least in part on user interaction in a virtual environment>") updating, based on the environmental
context, the output image to generate a second output image. (Ghosh, Paragraph [0020] “user profile 201 may store multiple personae or representations associated with the user"; [0023] “context engine 106 selects one of the user's representations based on the current context and causes the selected representation to be presented to the other party; [0041] "as the user's context changes, context engine 106 may select a different representation and cause the different representation to be displayed")
Ghosh a d Shechtman/He are analogous because all concern computer-implemented systems that present imagery (composite images or user representations) to other parties, where the presentation is controlled by contextual information and/or user input. Shechtman provide ways of generating an output image using generative models from multiple input images and presenting the resulting output image to a user. Ghosh provides a way of determining a user's current context from interactions with services in a virtual environment and then changing the representation that is presented based on that context. Therefore, it would have been obvious before the effective filing date to incorporate Ghosh's context-determination and representation-updating into Shechtman/He's image-generation framework such that, after an output image is generated for a user, the system determines an environmental context based at least in part on user interaction in a virtual environment and updates the output image based
on that context o generate and present a second output image. The motivation is to adapt the presented imagery to the user's current situation and environment, improving relevance and user experience, as discussed by Ghosh in determining "current context ... based on ... interactions" and selecting different representations in Paragraphs [0030], [0023] and [0041 ].
Regarding Claim 29, the combination of Shechtman, He, Cok and Ghosh teaches the
invention in Claim 28.
The combination further teaches the environmental context includes at least one of a professional environment, personal environment, family environment, or friendly environment (Ghosh, Paragraph [0015), "the context in which a user accesses a service impacts the user's behavior, preferences, and/or interests when accessing the service. The context includes the environment of the user (e.g., location of the user) <read on professional environment I personal environment>, time information, tasks being performed by the user <read on professional vs. personal activity> , whether the service accessed is associated with a trusted or non-trusted service provider, a social network of the user <read on friendly environment or family environment depending on social ties>]
Ghosh and Shechtman are analogous for the same reasons as in Claim 28:
all involve adapting computer-generated visual content based on context. Shechtman provide the mechanism for generating and presenting output images, while Ghosh supplies detailed categories of environmental context (work, home, family, social events). Therefore, it would have been obvious before the effective filing date to incorporate configure the environmental context in the combined system to include professional, personal, family, and friendly environments taught by Ghosh into modified invention of Shechtman such that system will further tailor the generated output images to users' varied real-word situations and social settings.
Regarding Claim 30, the combination of Shechtman, He, Cok and Ghosh teaches the
invention in Claim 28.
The combination further teaches executing a generative artificial intelligence (Al) model [[ to determine the theme ]] and generate the output image (Shechtman, Paragraph [0005] "methods for effectively generating composite images using a generative adversarial neural network."; [0044] "the im:age composite system can combine the geometric prediction neural network and the adversarial discrimination neural network to form a generative adversarial network (GAN)"; [0054) "Once trained, the image composite system can employ the generative adversarial network 100 to generate realistic composite images.")
But Shechtman does not explicitly disclose [[ executing a generative artificial intelligence (Al) model ]] to determine the theme.
However, He teaches executing a generative artificial intelligence (Al) model to
determine the theme and generate the output image. (He, Paragraph [0036) "Step 201 includes acquiring at least two frames of facial images extra4ted from a target video."; [0037], "The at least two frames of facial images are facial images of a same person."; Paragraph [0038] "Step 202 includes inputting the at least two frames of facial images into a pre-trained generative model to generate a single facial image."; it is noted facial images of a same person is a theme).
As explained in rejection of Claim 1, the obviousness for combining of theme of
He into Shechtman is provided above.
Response to Arguments
Applicant’s arguments with respect to claim 1, filed on 3/5/2026, with respect to rejection under 35 USC § 103 have been considered but are moot in view of the new ground(s) of rejection. It has now been taught by the combination of prior arts Shechtman, He and Cok.
In regard to Claims 2, 5-10 , 13-14, 19, 21-22, 24-30, they directly/indirectly depends on independent Claim 1. Applicant does not argue anything other than the independent claim 1. The limitations in those claims in conjunction with combination previously established as explained.
Applicant’s remarks state that “Claims 31 and 32 are newly presented.”. This statement is not accurate. Claims 31 and 32 were previously presented in the application and were rejected under 35 USC § 103 in the prior Office Action. Claims 31 and 32 remain pending and are addressed in the rejections set forth above.
It is noted that Applicant appears to have intended to state that Claims 33 and 34 are newly presented, as Claims 33 and 34 do not appear in the prior Office Action and are presented for the first time in the amendment filed on 03/05/2026. Claims 33 and 34 are treated accordingly as new claims and are addressed in the rejections set forth above.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 20030128877 A1 Method and system for processing images for themed imaging services
US 20060074861 A1 Reduction of seach ambiguity with multiple media references
US 20110029561 A1 Image similarity from disparate sources
US 8370282 B1 Image quality measures
US 20220414936 A1 Multimodal color variations using learned color distributions
Any inquiry concerning this communication or earlier communications from the examiner should be directed to YUJANG TSWEI whose telephone number is (571)272-6669. The examiner can normally be reached 8:30am-5:30pm EST.
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.
/YuJang Tswei/Primary Examiner, Art Unit 2614
Prosecution Insights