DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendments
The amendments to claims 1, 11, and 16 are accepted and entered.
Claims 1-20 are pending regarding this application.
Response to Arguments
Applicant’s arguments, see Remarks, filed 03/03/2026, with respect to the Claim Objection applied to claim 16 have been fully considered and are persuasive. The Claim Objection of claim 16 has been withdrawn.
Applicant’s arguments, see Remarks, filed 03/03/2026, with respect to the 112(b) Rejections applied to claims 1-20 have been fully considered and are persuasive. The 112(b) Rejections of claims 1-20 have been withdrawn.
Applicant’s arguments, see Remarks, filed 03/03/2026, with respect to the 112(a) Rejections applied to claims 1-20 have been fully considered and are persuasive. The 112(a) Rejections of claims 1-20 have been withdrawn.
Applicant's arguments filed on 03/03/2026 have been fully considered but they are not persuasive. Applicant argues
First, the Office contends that Kalish discloses dispatching the plurality of tasks to a plurality of tools, wherein the plurality of tools are associated with the centralized system, wherein the centralized system enables the plurality of tools to ... implement the plurality of tasks." Office Action, at 10-11. In support of this contention, the Office cites paragraph [0135] of Kalish, which discloses "[i]n this step, the module (reference number 708) determines the appropriate tool service for selecting or generating content and media objects." Id. at 11. The Office also asserts "Voiceover Generation, Text Placeholder Filling, Background Music Selection, along with the tools which carry out the tasks as noted in [Kalish's] para. [0134], etc... as noted in para. [0097-0099]) are interpreted as the tools as claimed in the claim language." Id. However, these portions of Kalish cited by the Office are not present in the Kalish Provisional, and are therefore not entitled to a January 20, 2023 priority date.
However, examiner disagrees with this assertion. Applicant further argues “nowhere does the Kalish Provisional disclose Voiceover Generation, Text Placeholder Filling, or Background Music Selection, which the Office alleges to be equivalent to the claimed plurality of tools. The Kalish Provisional only mentions the word "tools" in the following context "creating scenes, optionally generating new content using inter or external graphic multimedia tools." But creating scenes using tools is not equivalent to a machine learning model that dispatches tasks to at least a subset of a plurality of tools, as required by clarified claim 1”, however Kalish Provisional recites “creating scenes, optionally generating new content using inter or external graphic multimedia tools. Generate voiceover (using TTS, applying narrator and voice emotion (Friendly, excited, cheerful, advertisement). Generate text for all text placeholders. Select background music” (emphasis added) in para. [0045]-[0048]. This section is interpreted as equivalent to the cited section in para. [0097]-[0099] and sufficiently teaches “dispatching the plurality of tasks to a plurality of tools, wherein the plurality of tools are associated with the centralized system, wherein the centralized system enables the plurality of tools to ... implement the plurality of tasks”. See also FIG. 4 of Kalish Provisional which is a function of the centralized system as depicted in FIG. 1 of Kalish Provisional.
Additionally, applicant’s arguments regarding the following amended subject matter of claim 1: “dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data” (emphasis added) are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument. As a result, Patterson et al. (U.S. Publication No. 2021/0272599 A1), is included below in the prior art rejection of claim 1. See the 103 rejection of claims 1, 2, 5, 6, 7, 9, 10, 11, 12, 14, 15, 16, 17, 19, and 20 regarding this matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 2, 5, 6, 7, 9, 10, 11, 12, 14, 15, 16, 17, 19, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kalish et al. (U.S. Publication No. 2024/0249457 A1), hereinafter Kalish in view of Patterson et al. (U.S. Publication No. 2021/0272599 A1), hereinafter Patterson and Liu et al. (U.S. Publication No. 2022/0374637 A1), hereinafter Liu.
*Please note: Kalish et al. (U.S. Publication No. 2024/0249457 A1), hereinafter Kalish relies on a date of Provisional application (No. 63/480,714), filed on Jan 20, 2023. Since applicant has argued “many of the portions of Kalish relied on by the Office in the Office Action are not present in the Kalish Provisional”, additional citations have been included below regarding support for the inclusion of Kalish in the aforementioned provisional application. As such, all references to Provisional application (No. 63/480,714), filed on Jan 20, 2023, will be denoted as “Kalish Provisional”.
Regarding claim 1, Kalish teaches a method for orchestrating video creation using content creation tools by a machine learning model of a centralized system (Kalish teaches a centralized system in FIG. 1, see AI director bot module #700 which is interpreted as equivalent to the claimed machine learning model of the centralized system, see also para. [0158]. See support for this in FIG. 1 of Kalish Provisional), comprising:
receiving text by a machine learning model of the centralized system via a user interface, wherein the text indicates instructions for creating a video (Kalish teaches “video generation server 80 configured to receive entity/users' text, selection and customized data for generating relevant video parts based on pre-defined video templates or by using AI director module 700” in para. [0066]; here, the AI director module 700 is interpreted as the machine learning model of the centralized system as claimed in the claim language. See support for this in para. [0015], [0041]-[0042] and FIG. 1 of Kalish Provisional);
generating a script for the video based on the text by the machine learning model, wherein the script indicates a series of scenes in the video (Kalish teaches “based on the script derived from user input, the module (reference number 704) defines specific scenario parts or scenes” in para. [0133]; see also para. [0077] and [0132-0135]; See support for this in para. [0025], [0070], [0074]-[0076], and FIG. 1 of Kalish Provisional);
generating, by the machine learning model (Kalish teaches that the AI model carries out the below tasks as shown in FIGs. 7 and 8. See support for this in FIGs. 7 and 8 of Kalish Provisional), a plurality of tasks associated with creating the video based on the script (Kalish teaches that “for each part of the scenario, as delineated in the script, the module (reference number 706) determines several key elements. These include the layout style, context, content, the number of objects, types and properties of content objects, and the layout of video frames. It also establishes the sequence for displaying content, the functionality of objects, and provides options for object customization” in para. [0134]; see also para. [0124]; these key elements are interpreted as the tasks as described in the claim language. See support for this in para. [0075]-[0076] in Kalish Provisional);
dispatching the plurality of tasks to (Kalish teaches that “the module (reference number 708) determines the appropriate tool service for selecting or generating content and media objects” in para. [0135]; see also para. [0096]-[0099] which includes a description of the types of tools that may be implemented based on the analyzed script as shown in para. [0093-0094]; these tools (Voiceover Generation, Text Placeholder Filling, Background Music Selection, along with the tools which carry out the tasks as noted in para. [0134], etc… as noted in para. [0097-0099]) are interpreted as the tools as claimed in the claim language, wherein the tools are associated with the centralized system as depicted in FIG. 1. See support for this in para. [0041]-[0048] and FIG. 1 of Kalish Provisional);
collecting data indicating results of the plurality of tasks from (Kalish teaches “Template Selection and Customization (230)” and “Content Aggregation (240)” in para. [0094] and [0095], respectively. These two steps involve selecting “an appropriate video template or a combination of video scene templates” and “exploring and aggregating content from various internal and external sources” as shown in para. [0094] and [0095], respectively. These steps are interpreted as the steps in which the resulting data from the tool implementation is collected; see also para. [0135]. Support for this can be found in para. [0043]-[0044] of Kalish Provisional);
sending, by the machine learning model, the collected data (Kalish teaches “generating new video by implementing selected or new video template using aggregating content wherein the generated video complies with all analyzed requirements” in para. [0042]. See also para. [0133] wherein this process of determining which scenes to select/send is carried out by AI. Support for this can be found in para. [0049]-[0050] and para. [0076] of Kalish Provisional); and
displaying(Kalish teaches that “the user receives at least a part of the script, one or more audio parts, and one or more generated video segments. These elements are presented to the user for review and selection, providing a tangible representation of their initial instructions 330” in para. [0107]. Support for this can be found in para. [0055]-[0056] of Kalish Provisional).
Kalish fails to teach dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data (emphasis added).
However, Patterson teaches dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model (Patterson teaches “the models for processing and editing can be selected as particularly relevant to video on social media and/or professional editing tasks” in para. [0065], wherein “the processing system can employ machine learning models for spatial classification coupled with cinematic concept classification to enable narrative construction in output video” as shown in para. [0062]. Here, the teaching of a selection of relevant models is interpreted as equivalent to dispatching tasks to a subset of a plurality of tools (models)); and
sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data (Patterson teaches “operations within the embedding space are used to create output sequences of video” in para. [0153]. Patterson additionally teaches a process wherein “When the algorithms are integrated at a system level, first AttentionNet finds video clips that are out of the ordinary, then TrashNet filters the clips to those matching a desired category from the Processing System ontology, and finally the videos ordered by a Narrative Sequencer to produce the output edited video” as shown in para. [0134]. Here, the narrative sequencer is interpreted as equivalent to a tool (of a plurality of tools) used to generate the output video based on the data gathered and edited by AttentionNet and TrashNet. See also para. [0124]-[0131]. See also para. [0062] wherein this process uses machine learning models); and
and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data (Patterson teaches “the application transforms the content into a semantic embedding space and generates a rough-cut of a new video output including a series of video edits (e.g., temporal cuts, narrative transitions, trimmed video segments, etc.) at 306” in para. [0059]. In para. [0054], Patterson teaches that the machine learning model can produce the video to user such that the user can provide feedback which can subsequently be used to further refine/improve the machine learning model(s)).
Kalish and Patterson are both considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish to incorporate the teachings of Patterson and include “dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data”. The motivation for doing so would have been that “the system creates user interesting content in a much faster and easier method relative to any manual search, modifying search terms, editing and re-editing video. Ultimately, the result is an automated system that operates (video editing process) orders of magnitudes faster than conventional approaches, especially for tasks in which editors need to cut many actions together (sports highlights, music videos, etc.)”, as suggested by Patterson in para. [0113]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish with Patterson to obtain the invention specified in the above claim limitations.
Kalish and Patterson fail to teach a centralized system that enables the plurality of tools to simultaneously implement the plurality of tasks.
However, Liu teaches a centralized system (Liu, see FIG. 5 & para. [0072] which “can include a video generator that can perform one or more image editing, augmentation, modification, manipulation, or generation tasks as discussed and suggested herein” as shown in para. [0072]) that enables the plurality of tools to simultaneously implement the plurality of tasks (Liu teaches that “tasks [are] represented by a number of threads” in para. [0324] wherein “each individual thread and threads executing same instructions may be converged and executed in parallel for better efficiency” as further shown in para. [0324]; see also FIG. 27 and para. [0335]. See also FIG. 17A #1700).
Kalish, Patterson, and Liu are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson) to incorporate the teachings of Liu and include “wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks”. The motivation for doing so would have been to increase efficiency, as suggested by Liu in para. [0324]. See also para. [0335]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish and Patterson with Liu to obtain the invention specified in claim 1.
Regarding claim 2, Kalish, Patterson, and Liu teach the method of claim 1, further comprising:
performing a prompt engineering process (The prompt engineering process here is interpreted as a prompt (input) made by the user to trigger AI model learning; see also para. [0025] of the applicant’s specification which teaches the AI model learns from text descriptions regarding content creation tools) to enable the machine learning model to learn functions of the plurality of tools, application programming interfaces (APIs) of the plurality of tools, and parameters required by the plurality tools for implementing the plurality of task (Kalish teaches an AI model training process which learns from user input (this is interpreted as equivalent to the claimed prompt engineering process) wherein the AI model learns which templates/content/styles to use in the video generation process in order to best align the video with the user’s vision in para. [0126]-[0129]. This process of deciding how and when to utilize certain styles/tools/templates is interpreted as inherently equivalent to enabling the machine learning model to learn the functions of said tools, learning the parameters of the tools (i.e. when to use them), and the APIs of the tools. Support for this can be found in FIG. 7 of Kalish Provisional) (Patterson additionally teaches training the machine learning model in para. [0049]-[0050] and [0054] and FIGs. 1 and 2).
Regarding claim 5, Kalish, Patterson, and Liu teach the method of claim 1, further comprising:
receiving a video clip by the machine learning model via the user interface (Patterson teaches receiving a user input video by a machine learning model and processing the video in order to classify the “video segments into semantic embedding features, temporal sequence features, narrative features, etc” as shown in para. [0053]; Patterson further teaches taking in user text input, and generating a script based on the video analysis in para. [0087-0088]);
generating an analysis of the video by a video analysis tool associated with the centralized system, wherein the analysis indicates objects and themes detected in the video clip (Kalish teaches that the module (which is part of the centralized system as shown in FIG. 1) analyzes the image and identifies objects and themes in the image as shown in para. [0139-0140]) (Patterson teaches “the processing system can include a plurality of pre-trained models configured to recognize temporal events in input video, emotional events in input video, film-based events in input video, contextual events, and/or cinematic events” wherein “the machine learning models of the editing component 206 are configured to define a semantic embedding space that contains object, person, scene attributes, scene categories, and object categories” as shown in para. [0055]-[0056]); and
generating the script based on the analysis and the text by the machine learning model (Kalish teaches “the module generates a storyboard based on the analyzed image. This storyboard outlines the sequence of events or scenes that will be depicted in the video, ensuring a logical and engaging flow that is rooted in the content of the original image 906A” in para. [0141]; the storyboard here is interpreted as the script which is based on the process outlined in FIG. 9A; see also para. [0131] which discusses the storyboard in relation to the script and the user-provided text) (Patterson additionally teaches “the system is configured to generate a script based at least in part on natural language processing algorithms for conversation and narrative text generation” in para. [0088]).
Similar motivations as applied to claim 1 can be applied here to claim 5.
Regarding claim 6, Kalish, Patterson, and Liu teach the method of claim 1, further comprising:
receiving feedback information related to the video via the user interface, wherein the feedback information requests modifications to the video (Kalish teaches that “the user can enter additional instructions or edit previous ones. They also have the option to manually select more relevant media or utilize services like DALL-E-2 for media generation. Users can upload their own media or text, delete scenes, update the script, and, if desired, approve the final version. This step also includes enabling manual editing options, allowing for greater customization and personalization of the video content 350” in para. [0109]);
generating an updated script based on the feedback information, wherein the updated script indicates how the video is to be modified by the centralized system (Kalish teaches updating the script in the above citation from para. [0109], wherein the script acts as the instructions for how the video should be modified as taught in para. [0077] and [0132]); and
generating the modified video based at least in part on the updated script (Kalish teaches a Final Video Segment Selection (360) process in which the user can see the final version of the video segment containing all modifications and select the modified video if it is determined to be acceptable in para. [0111]).
Regarding claim 7, Kalish, Patterson, and Liu teach the method of claim 1, further comprising:
compiling the collected data indicating results of the plurality of tasks (Kalish teaches “aggregating content from various internal and external sources” in para. [0095]; this aggregated content is interpreted as equivalent to the compiled data as taught in the claim language); and
transmitting the compiled data to a video creation tool associated with the centralized system for generating the video (Kalish teaches “Final Video Generation (260): The module generates the new video by implementing the selected template(s) or a newly created video template” in para. [0102]; the module 700 is associated with the centralized system as shown in FIG. 1).
Regarding claim 9, Kalish, Patterson, and Liu teach the method of claim 1, wherein the plurality of tools comprises
a video editing tool (Kalish teaches that the process can include “facilitates … the editing of pre-made videos, such as cutting relevant parts or changing properties to better suit the script” in para. [0135]),
a music recommendation tool (Kalish teaches that the video can comprise music wherein the Content Aggregation step (240) can include “Background Music Selection: Selecting suitable background music to complement the video's mood and enhance the viewer's experience” as noted in para. [0099]; this selected music is interpreted as the music recommendation tool as taught in the claim language),
an image searching tool configured to search images based on a user input (Kalish teaches searching images based on input by the user in para. [0138-0142]; see also para. [0095] which describes searching through content, which may be images, in order to aggregate content according to the user’s input preferences), and
a text-to-speech tool configured to generate speech audio based on an input text (Kalish teaches that the video can comprise speech audio wherein the Content Aggregation step (240) can include “Voiceover Generation: A voiceover is generated using text-to-speech technology. The module applies appropriate narrators and voice emotions (e.g., friendly, excited, cheerful, advertisement style) to align with the video's tone” as noted in para. [0097], wherein the audio is based in part on the analysis of the user instructions as noted in para. [0095]).
Regarding claim 10, Kalish, Patterson, and Liu teach the method of claim 1,
wherein the video comprises
images (Kalish teaches that the video may include images in the Content Aggregation step (240) in para. [0095]; see also FIG. 9A and para. [0137]-[0142]),
music (Kalish teaches that the video can comprise music wherein the Content Aggregation step (240) can include “Background Music Selection: Selecting suitable background music to complement the video's mood and enhance the viewer's experience” as noted in para. [0099]),
speech audio (Kalish teaches that the video can comprise speech audio wherein the Content Aggregation step (240) can include “Voiceover Generation: A voiceover is generated using text-to-speech technology. The module applies appropriate narrators and voice emotions (e.g., friendly, excited, cheerful, advertisement style) to align with the video's tone” as noted in para. [0097]), and
text (Kalish teaches that the video can comprise text wherein the Content Aggregation step (240) can include “Text Placeholder Filling: The module generates text for all text placeholders in the video, ensuring consistency and relevance to the video's content” as noted in para. [0098]).
Regarding claim 11, Kalish teaches a system (Kalish, FIG. 1) for generating videos using a centralized system (Kalish teaches a centralized system in FIG. 1, see also para. [0158]. See support for this in FIG. 1 of Kalish Provisional), comprising:
at least one processor; and at least one memory comprising computer-readable instructions that upon execution by the at least one processor cause the system to perform operations comprising (Kalish teaches a method for generating videos which is “implemented by one or more processors operatively coupled to a non-transitory computer readable storage device, on which are stored modules of instruction code that when executed cause the one or more processors to perform the steps of …” in para. [0014]; here, the non-transitory crm is interpreted as equivalent to memory comprising computer readable instructions. Support for this can be found in para. [0002]-[0004] of Kalish Provisional):
receiving text by a machine learning model of the centralized system via a user interface, wherein the text indicates instructions for creating a video (Kalish teaches “video generation server 80 configured to receive entity/users' text, selection and customized data for generating relevant video parts based on pre-defined video templates or by using AI director module 700” in para. [0066]; here, the AI director module 700 is interpreted as the machine learning model of the centralized system as claimed in the claim language. See support for this in para. [0015], [0041]-[0042] and FIG. 1 of Kalish Provisional);
generating a script for the video based on the text by the machine learning model, wherein the script indicates a series of scenes in the video (Kalish teaches “based on the script derived from user input, the module (reference number 704) defines specific scenario parts or scenes” in para. [0133]; see also para. [0077] and [0132-0135]; See support for this in para. [0025], [0070], [0074]-[0076], and FIG. 1 of Kalish Provisional);
generating, by the machine learning model (Kalish teaches that the AI model carries out the below tasks as shown in FIGs. 7 and 8. See support for this in FIGs. 7 and 8 of Kalish Provisional), a plurality of tasks associated with creating the video based on the script (Kalish teaches that “for each part of the scenario, as delineated in the script, the module (reference number 706) determines several key elements. These include the layout style, context, content, the number of objects, types and properties of content objects, and the layout of video frames. It also establishes the sequence for displaying content, the functionality of objects, and provides options for object customization” in para. [0134]; see also para. [0124]; these key elements are interpreted as the tasks as described in the claim language. See support for this in para. [0075]-[0076] in Kalish Provisional);
dispatching the plurality of tasks to (Kalish teaches that “the module (reference number 708) determines the appropriate tool service for selecting or generating content and media objects” in para. [0135]; see also para. [0096]-[0099] which includes a description of the types of tools that may be implemented based on the analyzed script as shown in para. [0093-0094]; these tools (Voiceover Generation, Text Placeholder Filling, Background Music Selection, along with the tools which carry out the tasks as noted in para. [0134], etc… as noted in para. [0097-0099]) are interpreted as the tools as claimed in the claim language, wherein the tools are associated with the centralized system as depicted in FIG. 1. See support for this in para. [0041]-[0048] and FIG. 1 of Kalish Provisional);
collecting data indicating results of the plurality of tasks from (Kalish teaches “Template Selection and Customization (230)” and “Content Aggregation (240)” in para. [0094] and [0095], respectively. These two steps involve selecting “an appropriate video template or a combination of video scene templates” and “exploring and aggregating content from various internal and external sources” as shown in para. [0094] and [0095], respectively. These steps are interpreted as the steps in which the resulting data from the tool implementation is collected; see also para. [0135]. Support for this can be found in para. [0043]-[0044] of Kalish Provisional);
sending, by the machine learning model, the collected data (Kalish teaches “generating new video by implementing selected or new video template using aggregating content wherein the generated video complies with all analyzed requirements” in para. [0042]. See also para. [0133] wherein this process of determining which scenes to select/send is carried out by AI. Support for this can be found in para. [0049]-[0050] and para. [0076] of Kalish Provisional); and
displaying(Kalish teaches that “the user receives at least a part of the script, one or more audio parts, and one or more generated video segments. These elements are presented to the user for review and selection, providing a tangible representation of their initial instructions 330” in para. [0107]. Support for this can be found in para. [0055]-[0056] of Kalish Provisional).
Kalish fails to teach dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data (emphasis added).
However, Patterson teaches dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model (Patterson teaches “the models for processing and editing can be selected as particularly relevant to video on social media and/or professional editing tasks” in para. [0065], wherein “the processing system can employ machine learning models for spatial classification coupled with cinematic concept classification to enable narrative construction in output video” as shown in para. [0062]. Here, the teaching of a selection of relevant models is interpreted as equivalent to dispatching tasks to a subset of a plurality of tools (models)); and
sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data (Patterson teaches “operations within the embedding space are used to create output sequences of video” in para. [0153]. Patterson additionally teaches a process wherein “When the algorithms are integrated at a system level, first AttentionNet finds video clips that are out of the ordinary, then TrashNet filters the clips to those matching a desired category from the Processing System ontology, and finally the videos ordered by a Narrative Sequencer to produce the output edited video” as shown in para. [0134]. Here, the narrative sequencer is interpreted as equivalent to a tool (of a plurality of tools) used to generate the output video based on the data gathered and edited by AttentionNet and TrashNet. See also para. [0124]-[0131]. See also para. [0062] wherein this process uses machine learning models); and
and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data (Patterson teaches “the application transforms the content into a semantic embedding space and generates a rough-cut of a new video output including a series of video edits (e.g., temporal cuts, narrative transitions, trimmed video segments, etc.) at 306” in para. [0059]. In para. [0054], Patterson teaches that the machine learning model can produce the video to user such that the user can provide feedback which can subsequently be used to further refine/improve the machine learning model(s)).
Kalish and Patterson are both considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish to incorporate the teachings of Patterson and include “dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data”. The motivation for doing so would have been that “the system creates user interesting content in a much faster and easier method relative to any manual search, modifying search terms, editing and re-editing video. Ultimately, the result is an automated system that operates (video editing process) orders of magnitudes faster than conventional approaches, especially for tasks in which editors need to cut many actions together (sports highlights, music videos, etc.)”, as suggested by Patterson in para. [0113]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish with Patterson to obtain the invention specified in the above claim limitations.
Kalish and Patterson fail to teach a centralized system that enables the plurality of tools to simultaneously implement the plurality of tasks.
However, Liu teaches a centralized system (Liu, see FIG. 5 & para. [0072] which “can include a video generator that can perform one or more image editing, augmentation, modification, manipulation, or generation tasks as discussed and suggested herein” as shown in para. [0072]) that enables the plurality of tools to simultaneously implement the plurality of tasks (Liu teaches that “tasks [are] represented by a number of threads” in para. [0324] wherein “each individual thread and threads executing same instructions may be converged and executed in parallel for better efficiency” as further shown in para. [0324]; see also FIG. 27 and para. [0335]. See also FIG. 17A #1700).
Kalish, Patterson, and Liu are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson) to incorporate the teachings of Liu and include “wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks”. The motivation for doing so would have been to increase efficiency, as suggested by Liu in para. [0324]. See also para. [0335]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish and Patterson with Liu to obtain the invention specified in claim 11.
Regarding claim 12, Kalish, Patterson, and Liu teach the system of claim 11, the operations further comprising:
performing a prompt engineering process (The prompt engineering process here is interpreted as a prompt (input) made by the user to trigger AI model learning; see also para. [0025] of the applicant’s specification which teaches the AI model learns from text descriptions regarding content creation tools) to enable the machine learning model to learn functions of the plurality of tools, application programming interfaces (APIs) of the plurality of tools, and parameters required by the plurality tools for implementing the plurality of task (Kalish teaches an AI model training process which learns from user input (this is interpreted as equivalent to the claimed prompt engineering process) wherein the AI model learns which templates/content/styles to use in the video generation process in order to best align the video with the user’s vision in para. [0126]-[0129]. This process of deciding how and when to utilize certain styles/tools/templates is interpreted as inherently equivalent to enabling the machine learning model to learn the functions of said tools, learning the parameters of the tools (i.e. when to use them), and the APIs of the tools. Support for this can be found in FIG. 7 of Kalish Provisional) (Patterson additionally teaches training the machine learning model in para. [0049]-[0050] and [0054] and FIGs. 1 and 2).
Regarding claim 14, Kalish, Patterson, and Liu teach the system of claim 11, the operations further comprising:
receiving a video clip by the machine learning model via the user interface (Patterson teaches receiving a user input video by a machine learning model and processing the video in order to classify the “video segments into semantic embedding features, temporal sequence features, narrative features, etc” as shown in para. [0053]; Patterson further teaches taking in user text input, and generating a script based on the video analysis in para. [0087-0088]);
generating an analysis of the video by a video analysis tool associated with the centralized system, wherein the analysis indicates objects and themes detected in the video clip (Kalish teaches that the module (which is part of the centralized system as shown in FIG. 1) analyzes the image and identifies objects and themes in the image as shown in para. [0139-0140]) (Patterson teaches “the processing system can include a plurality of pre-trained models configured to recognize temporal events in input video, emotional events in input video, film-based events in input video, contextual events, and/or cinematic events” wherein “the machine learning models of the editing component 206 are configured to define a semantic embedding space that contains object, person, scene attributes, scene categories, and object categories” as shown in para. [0055]-[0056]); and
generating the script based on the analysis and the text by the machine learning model (Kalish teaches “the module generates a storyboard based on the analyzed image. This storyboard outlines the sequence of events or scenes that will be depicted in the video, ensuring a logical and engaging flow that is rooted in the content of the original image 906A” in para. [0141]; the storyboard here is interpreted as the script which is based on the process outlined in FIG. 9A; see also para. [0131] which discusses the storyboard in relation to the script and the user-provided text) (Patterson additionally teaches “the system is configured to generate a script based at least in part on natural language processing algorithms for conversation and narrative text generation” in para. [0088]).
Similar motivations as applied to claim 11 can be applied here to claim 14.
Regarding claim 15, Kalish, Patterson, and Liu teach the system of claim 11, the operations further comprising:
receiving feedback information related to the video via the user interface, wherein the feedback information requests modifications to the video (Kalish teaches that “the user can enter additional instructions or edit previous ones. They also have the option to manually select more relevant media or utilize services like DALL-E-2 for media generation. Users can upload their own media or text, delete scenes, update the script, and, if desired, approve the final version. This step also includes enabling manual editing options, allowing for greater customization and personalization of the video content 350” in para. [0109]);
generating an updated script based on the feedback information, wherein the updated script indicates how the video is to be modified by the centralized system (Kalish teaches updating the script in the above citation from para. [0109], wherein the script acts as the instructions for how the video should be modified as taught in para. [0077] and [0132]); and
generating the modified video based at least in part on the updated script (Kalish teaches a Final Video Segment Selection (360) process in which the user can see the final version of the video segment containing all modifications and select the modified video if it is determined to be acceptable in para. [0111]).
Regarding claim 16, Kalish teaches a non-transitory computer-readable storage medium, storing computer-readable instructions that upon execution by a processor cause the processor to implement operations (Kalish teaches a method for generating videos which is “implemented by one or more processors operatively coupled to a non-transitory computer readable storage device, on which are stored modules of instruction code that when executed cause the one or more processors to perform the steps of …” in para. [0014]. Support for this can be found in para. [0002]-[0004] in Kalish Provisional), the operation comprising:
receiving text by a machine learning model of the centralized system via a user interface, wherein the text indicates instructions for creating a video (Kalish teaches “video generation server 80 configured to receive entity/users' text, selection and customized data for generating relevant video parts based on pre-defined video templates or by using AI director module 700” in para. [0066]; here, the AI director module 700 is interpreted as the machine learning model of the centralized system as claimed in the claim language. See support for this in para. [0015], [0041]-[0042] and FIG. 1 of Kalish Provisional);
generating a script for the video based on the text by the machine learning model, wherein the script indicates a series of scenes in the video (Kalish teaches “based on the script derived from user input, the module (reference number 704) defines specific scenario parts or scenes” in para. [0133]; see also para. [0077] and [0132-0135]; See support for this in para. [0025], [0070], [0074]-[0076], and FIG. 1 of Kalish Provisional);
generating, by the machine learning model (Kalish teaches that the AI model carries out the below tasks as shown in FIGs. 7 and 8. See support for this in FIGs. 7 and 8 of Kalish Provisional), a plurality of tasks associated with creating the video based on the script (Kalish teaches that “for each part of the scenario, as delineated in the script, the module (reference number 706) determines several key elements. These include the layout style, context, content, the number of objects, types and properties of content objects, and the layout of video frames. It also establishes the sequence for displaying content, the functionality of objects, and provides options for object customization” in para. [0134]; see also para. [0124]; these key elements are interpreted as the tasks as described in the claim language. See support for this in para. [0075]-[0076] in Kalish Provisional);
dispatching the plurality of tasks to plurality of tasks (Kalish teaches that “the module (reference number 708) determines the appropriate tool service for selecting or generating content and media objects” in para. [0135]; see also para. [0096]-[0099] which includes a description of the types of tools that may be implemented based on the analyzed script as shown in para. [0093-0094]; these tools (Voiceover Generation, Text Placeholder Filling, Background Music Selection, along with the tools which carry out the tasks as noted in para. [0134], etc… as noted in para. [0097-0099]) are interpreted as the tools as claimed in the claim language, wherein the tools are associated with the centralized system as depicted in FIG. 1. See support for this in para. [0041]-[0048] and FIG. 1 of Kalish Provisional);
collecting data indicating results of the plurality of tasks from (Kalish teaches “Template Selection and Customization (230)” and “Content Aggregation (240)” in para. [0094] and [0095], respectively. These two steps involve selecting “an appropriate video template or a combination of video scene templates” and “exploring and aggregating content from various internal and external sources” as shown in para. [0094] and [0095], respectively. These steps are interpreted as the steps in which the resulting data from the tool implementation is collected; see also para. [0135]. Support for this can be found in para. [0043]-[0044] of Kalish Provisional);
sending, by the machine learning model, the collected data (Kalish teaches “generating new video by implementing selected or new video template using aggregating content wherein the generated video complies with all analyzed requirements” in para. [0042]. See also para. [0133] wherein this process of determining which scenes to select/send is carried out by AI. Support for this can be found in para. [0049]-[0050] and para. [0076] of Kalish Provisional); and
displaying(Kalish teaches that “the user receives at least a part of the script, one or more audio parts, and one or more generated video segments. These elements are presented to the user for review and selection, providing a tangible representation of their initial instructions 330” in para. [0107]. Support for this can be found in para. [0055]-[0056] of Kalish Provisional).
Kalish fails to teach dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data (emphasis added).
However, Patterson teaches dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model (Patterson teaches “the models for processing and editing can be selected as particularly relevant to video on social media and/or professional editing tasks” in para. [0065], wherein “the processing system can employ machine learning models for spatial classification coupled with cinematic concept classification to enable narrative construction in output video” as shown in para. [0062]. Here, the teaching of a selection of relevant models is interpreted as equivalent to dispatching tasks to a subset of a plurality of tools (models)); and
sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data (Patterson teaches “operations within the embedding space are used to create output sequences of video” in para. [0153]. Patterson additionally teaches a process wherein “When the algorithms are integrated at a system level, first AttentionNet finds video clips that are out of the ordinary, then TrashNet filters the clips to those matching a desired category from the Processing System ontology, and finally the videos ordered by a Narrative Sequencer to produce the output edited video” as shown in para. [0134]. Here, the narrative sequencer is interpreted as equivalent to a tool (of a plurality of tools) used to generate the output video based on the data gathered and edited by AttentionNet and TrashNet. See also para. [0124]-[0131]. See also para. [0062] wherein this process uses machine learning models); and
and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data (Patterson teaches “the application transforms the content into a semantic embedding space and generates a rough-cut of a new video output including a series of video edits (e.g., temporal cuts, narrative transitions, trimmed video segments, etc.) at 306” in para. [0059]. In para. [0054], Patterson teaches that the machine learning model can produce the video to user such that the user can provide feedback which can subsequently be used to further refine/improve the machine learning model(s)).
Kalish and Patterson are both considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish to incorporate the teachings of Patterson and include “dispatching the plurality of tasks to at least a subset of a plurality of tools by the machine learning model; wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks; sending, by the machine learning model, the collected data to one of the plurality of tools to cause the one of the plurality of tools to generate the video based on the collected data; and displaying, by the machine learning model, information on the user interface for accessing the video generated based on the collected data”. The motivation for doing so would have been that “the system creates user interesting content in a much faster and easier method relative to any manual search, modifying search terms, editing and re-editing video. Ultimately, the result is an automated system that operates (video editing process) orders of magnitudes faster than conventional approaches, especially for tasks in which editors need to cut many actions together (sports highlights, music videos, etc.)”, as suggested by Patterson in para. [0113]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish with Patterson to obtain the invention specified in the above claim limitations.
Kalish and Patterson fail to teach a centralized system that enables the plurality of tools to simultaneously implement the plurality of tasks.
However, Liu teaches a centralized system (Liu, see FIG. 5 & para. [0072] which “can include a video generator that can perform one or more image editing, augmentation, modification, manipulation, or generation tasks as discussed and suggested herein” as shown in para. [0072]) that enables the plurality of tools to simultaneously implement the plurality of tasks (Liu teaches that “tasks [are] represented by a number of threads” in para. [0324] wherein “each individual thread and threads executing same instructions may be converged and executed in parallel for better efficiency” as further shown in para. [0324]; see also FIG. 27 and para. [0335]. See also FIG. 17A #1700).
Kalish, Patterson, and Liu are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson) to incorporate the teachings of Liu and include “wherein the centralized system enables the plurality of tools to simultaneously implement the plurality of tasks”. The motivation for doing so would have been to increase efficiency, as suggested by Liu in para. [0324]. See also para. [0335]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish and Patterson with Liu to obtain the invention specified in claim 16.
Regarding claim 17, Kalish, Patterson, and Liu teach the non-transitory computer-readable storage medium of claim 16, the operations further comprising:
performing a prompt engineering process (The prompt engineering process here is interpreted as a prompt (input) made by the user to trigger AI model learning; see also para. [0025] of the applicant’s specification which teaches the AI model learns from text descriptions regarding content creation tools) to enable the machine learning model to learn functions of the plurality of tools, application programming interfaces (APIs) of the plurality of tools, and parameters required by the plurality tools for implementing the plurality of task (Kalish teaches an AI model training process which learns from user input (this is interpreted as equivalent to the claimed prompt engineering process) wherein the AI model learns which templates/content/styles to use in the video generation process in order to best align the video with the user’s vision in para. [0126]-[0129]. This process of deciding how and when to utilize certain styles/tools/templates is interpreted as inherently equivalent to enabling the machine learning model to learn the functions of said tools, learning the parameters of the tools (i.e. when to use them), and the APIs of the tools. Support for this can be found in FIG. 7 of Kalish Provisional) (Patterson additionally teaches training the machine learning model in para. [0049]-[0050] and [0054] and FIGs. 1 and 2).
Regarding claim 19, Kalish, Patterson, and Liu teach the non-transitory computer-readable storage medium of claim 16, the operations further comprising:
receiving a video clip by the machine learning model via the user interface (Patterson teaches receiving a user input video by a machine learning model and processing the video in order to classify the “video segments into semantic embedding features, temporal sequence features, narrative features, etc” as shown in para. [0053]; Patterson further teaches taking in user text input, and generating a script based on the video analysis in para. [0087-0088]);
generating an analysis of the video by a video analysis tool associated with the centralized system, wherein the analysis indicates objects and themes detected in the video clip (Kalish teaches that the module (which is part of the centralized system as shown in FIG. 1) analyzes the image and identifies objects and themes in the image as shown in para. [0139-0140]) (Patterson teaches “the processing system can include a plurality of pre-trained models configured to recognize temporal events in input video, emotional events in input video, film-based events in input video, contextual events, and/or cinematic events” wherein “the machine learning models of the editing component 206 are configured to define a semantic embedding space that contains object, person, scene attributes, scene categories, and object categories” as shown in para. [0055]-[0056]); and
generating the script based on the analysis and the text by the machine learning model (Kalish teaches “the module generates a storyboard based on the analyzed image. This storyboard outlines the sequence of events or scenes that will be depicted in the video, ensuring a logical and engaging flow that is rooted in the content of the original image 906A” in para. [0141]; the storyboard here is interpreted as the script which is based on the process outlined in FIG. 9A; see also para. [0131] which discusses the storyboard in relation to the script and the user-provided text) (Patterson additionally teaches “the system is configured to generate a script based at least in part on natural language processing algorithms for conversation and narrative text generation” in para. [0088]).
Similar motivations as applied to claim 16 can be applied here to claim 19.
Regarding claim 20, Kalish, Patterson, and Liu teach the non-transitory computer-readable storage medium of claim 16, the operations further comprising:
receiving feedback information related to the video via the user interface, wherein the feedback information requests modifications to the video (Kalish teaches that “the user can enter additional instructions or edit previous ones. They also have the option to manually select more relevant media or utilize services like DALL-E-2 for media generation. Users can upload their own media or text, delete scenes, update the script, and, if desired, approve the final version. This step also includes enabling manual editing options, allowing for greater customization and personalization of the video content 350” in para. [0109]. Support for this can be found in para. [0057] of Kalish Provisional);
generating an updated script based on the feedback information, wherein the updated script indicates how the video is to be modified by the centralized system (Kalish teaches updating the script in the above citation from para. [0109], wherein the script acts as the instructions for how the video should be modified as taught in para. [0077] and [0132]); and
generating the modified video based at least in part on the updated script (Kalish teaches a Final Video Segment Selection (360) process in which the user can see the final version of the video segment containing all modifications and select the modified video if it is determined to be acceptable in para. [0111]).
Claims 3, 4, 13, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Kalish et al. (U.S. Publication No. 2024/0249457 A1), hereinafter Kalish in view of Patterson et al. (U.S. Publication No. 2021/0272599 A1), hereinafter Patterson, Liu et al. (U.S. Publication No. 2022/0374637 A1), hereinafter Liu, and Sievert et al. (U.S. Publication No. 2016/0196852 A1), hereinafter Sievert.
Regarding claim 3, Kalish, Patterson, and Liu teach the method of claim 1, further comprising:
generating a (Kalish teaches that “the video file format of digital media container 300 is comprised of video or audio data 302 and meta data 304. The meta data comprises at least video ID or a link 306 and/or optionally partial or full video generation instructions 308 and/or customized parameters 310” in para. [0070]; here, the metadata which comprises the full video generation instructions and customized parameters is interpreted as equivalent to the files corresponding to the tasks as claimed in the claim language).
While Kalish teaches the above process in the context of a singular file, Kalish, Patterson, and Liu fail to teach a plurality of files each corresponding to the plurality of tasks.
However, Sievert teaches the above process in the context of a plurality of files each corresponding to the plurality of tasks (Sievert teaches that “the edit decision list encodes a series of flags (or sequencing files) that describe tasks to generate the edited video” in para. [0050] wherein “the edit decision list includes transcoding instructions to apply to a video or a portion of the video” as shown in para. [0055]).
Kalish, Patterson, Liu, and Sievert are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson and Liu) to incorporate the teachings of Sievert and include the above process in the context of a plurality of files each corresponding to the plurality of tasks. The motivation for doing so would have been to “to create an edited HD video” which contains the edits associated with the transcoded tasks, as suggested by Sievert in para. [0055]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish, Patterson, and Liu with Sievert to obtain the invention specified in claim 3.
Regarding claim 4, Kalish, Liu, Patterson, and Sievert teach the method of claim 3, further comprising:
transmitting the plurality of files to the plurality of tools via application programming interfaces (APIs) of the plurality of tools for simultaneously implementing the plurality of tasks by the plurality of tools (As shown in claim 3, Sievert teaches the plurality of files in the context of carrying out tasks. However, Kalish teaches the metadata file containing instructions for creating the videos in para. [0079], wherein the instructions (i.e. tasks) are provided to the tools in order to create a template which closely matches the user’s instructions as shown in para. [0094]). Similar motivations as applied to claim 3 can be applied here.
Regarding claim 13, Kalish, Patterson, and Liu teach the system of claim 11, the operations further comprising:
generating a (Kalish teaches that “the video file format of digital media container 300 is comprised of video or audio data 302 and meta data 304. The meta data comprises at least video ID or a link 306 and/or optionally partial or full video generation instructions 308 and/or customized parameters 310” in para. [0070]; here, the metadata which comprises the full video generation instructions and customized parameters is interpreted as equivalent to the files corresponding to the tasks as claimed in the claim language); and
transmitting the (Kalish teaches the metadata file containing instructions for creating the videos in para. [0079], wherein the instructions (i.e. tasks) are provided to the tools in order to create a template which closely matches the user’s instructions as shown in para. [0094]).
While Kalish teaches the above processes in the context of a singular file, Kalish, Patterson, and Liu fail to teach a plurality of files each corresponding to the plurality of tasks.
However, Sievert teaches the above processes in the context of a plurality of files each corresponding to the plurality of tasks (Sievert teaches that “the edit decision list encodes a series of flags (or sequencing files) that describe tasks to generate the edited video” in para. [0050] wherein “the edit decision list includes transcoding instructions to apply to a video or a portion of the video” as shown in para. [0055]).
Kalish, Patterson, Liu, and Sievert are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson and Liu) to incorporate the teachings of Sievert and include the above process in the context of a plurality of files each corresponding to the plurality of tasks. The motivation for doing so would have been to “to create an edited HD video” which contains the edits associated with the transcoded tasks, as suggested by Sievert in para. [0055]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish, Patterson, and Liu with Sievert to obtain the invention specified in claim 13.
Regarding claim 18, Kalish, Patterson, and Liu teach the non-transitory computer-readable storage medium of claim 16, the operations further comprising:
generating a (Kalish teaches that “the video file format of digital media container 300 is comprised of video or audio data 302 and meta data 304. The meta data comprises at least video ID or a link 306 and/or optionally partial or full video generation instructions 308 and/or customized parameters 310” in para. [0070]; here, the metadata which comprises the full video generation instructions and customized parameters is interpreted as equivalent to the files corresponding to the tasks as claimed in the claim language); and
transmitting the (Kalish teaches the metadata file containing instructions for creating the videos in para. [0079], wherein the instructions (i.e. tasks) are provided to the tools in order to create a template which closely matches the user’s instructions as shown in para. [0094]).
While Kalish teaches the above processes in the context of a singular file, Kalish, Patterson, and Liu fail to teach a plurality of files each corresponding to the plurality of tasks.
However, Sievert teaches the above processes in the context of a plurality of files each corresponding to the plurality of tasks (Sievert teaches that “the edit decision list encodes a series of flags (or sequencing files) that describe tasks to generate the edited video” in para. [0050] wherein “the edit decision list includes transcoding instructions to apply to a video or a portion of the video” as shown in para. [0055]).
Kalish, Patterson, Liu, and Sievert are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson and Liu) to incorporate the teachings of Sievert and include the above process in the context of a plurality of files each corresponding to the plurality of tasks. The motivation for doing so would have been to “to create an edited HD video” which contains the edits associated with the transcoded tasks, as suggested by Sievert in para. [0055]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish, Patterson, and Liu with Sievert to obtain the invention specified in claim 18.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Kalish et al. (U.S. Publication No. 2024/0249457 A1), hereinafter Kalish in view of Patterson et al. (U.S. Publication No. 2021/0272599 A1), hereinafter Patterson, Liu et al. (U.S. Publication No. 2022/0374637 A1), hereinafter Liu, and Boyle et al. (U.S. Publication No. 2015/0208023 A1), hereinafter Boyle.
Regarding claim 8, Kalish, Patterson, and Liu teach the method of claim 1.
While Kalish , Patterson, and Liu teach the centralized system as claimed in claim 1, and Liu teaches a machine model which carries out the instructions of the method as taught by Liu in para. [0246] and [0353], Kalish, Patterson, and Liu fail to teach automatically uploading the video to a server based on an instruction provided by the machine learning model to an uploading tool associated with the centralized system.
However, Boyle teaches automatically uploading the video to a server based on an instruction provided by the machine learning model (Boyle teaches FIG. 2 which illustrates the machine learning process in which a video is uploaded to the host computer (server) according to the neural network as shown in para. [0014]) to an uploading tool associated with the centralized system (Boyle additionally teaches that this uploading process may occur within a central computer system as shown in FIG. 1. Since the process of uploading a video to a server inherently involves some form of uploading process, it is inherent that an uploading tool would have to be present in order to carry out this process).
Kalish, Patterson, Liu, and Boyle are all considered to be analogous to the claimed invention because they are in the same field of generating videos based on user input. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the teachings of Kalish (as modified by Patterson and Liu) to incorporate the teachings of Boyle and include “automatically uploading the video to a server based on an instruction provided by the machine learning model to an uploading tool associated with the centralized system”. The motivation for doing so would have been that “the neural network of the present invention may be used to improve the automated tracking, automatic video recording, and/or automatic video production algorithm. The system may also expand the scope of what it can learn in order to automatically improve recording and editing functions”, as suggested by Boyle in para. [0011]. Therefore, it would have been obvious to one of ordinary skill at the time the invention was filed to combine Kalish, Patterson, and Liu with Boyle to obtain the invention specified in claim 8.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Bombach et al. (U.S. Publication No. 2024/0378850 A1) teaches a method of generating modified videos from original videos using a machine learning model.
Parasnis et al. (U.S. Patent No. 11809688 B1) teaches a method of generating videos using Generative Artificial Intelligence tools.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office
action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the
extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from
the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH
shortened statutory period, then the shortened statutory period will expire on the date the advisory
action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing
date of the advisory action. In no event, however, will the statutory period for reply expire later than
SIX MONTHS from the date of this final action.
Contact
Any inquiry concerning this communication or earlier communications from the examiner
should be directed to KYLA G ALLEN whose telephone number is (703)756-5315. The examiner can
normally be reached M-F 7:30am - 4:30pm 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,
John Villecco can be reached on (571) 272-7319. 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.
/Kyla Guan-Ping Tiao Allen/
Examiner, Art Unit 2661
/JOHN VILLECCO/Supervisory Patent Examiner, Art Unit 2661
Prosecution Insights