System and method for restricting video content

Abstract

Methods, systems, and computer program products are provided for restricting video content, including displaying, via an interface, an output of a portion of a video recommended for preview to facilitate restricting the portion of the video for one or more audience groups during playback; an indication is received via the interface and for the user profile whether the portion of the video is restricted for the one or more audience groups during playback, and the indication is stored for restricting use in the portion of the video for the one or more audience groups during playback.

Description

Background Art

[0001] Video streaming services are provided to allow users to stream videos from a server to personal devices, such as computers, televisions, phones, tablets, etc. Video streaming services typically require users to use the service to establish a user profile, which can be subscription-based, in order to access and play videos. Video content may have ratings representing the content in the video, such as restricted (R), parental guidance (PG), general (G), etc., and / or may have labels indicating whether the content has violence, adult scenes, offensive language, smoking, etc. These ratings and labels can help users decide what video content they want to play for themselves and / or family members. In some examples, a user profile can set a family member profile that can limit the content that can be played, which can be based on these ratings or labels. However, ratings and labels may not accurately represent all situations in the video content, and / or ratings and labels may prevent family members from watching video content that is generally not offensive (except for a small portion of one or more scenes). Summary of the invention

[0002] One or more aspects are briefly summarized below to provide a basic understanding of these aspects. The content of the present invention is not an extensive overview of all expected aspects, and is not intended to identify the key or important elements of all aspects, nor is it intended to describe the scope of any or all aspects. Its only purpose is to present some concepts of one or more aspects in a simplified form as a preface to a more detailed description later.

[0003] In an example, a computer-implemented method for limiting video content is provided, the method comprising displaying, via an interface, an output of a portion of a video recommended for preview to facilitate limiting the portion of the video for one or more audience groups during playback; receiving, via the interface and for a user profile, an indication as to whether to limit a portion of the video for one or more audience groups during playback; and storing the indication for use in limiting the portion of the video for one or more audience groups during playback.

[0004] In another example, a device for limiting video content is provided, the device comprising a memory storing instructions and at least one processor coupled to the memory. The at least one processor is configured to execute the instructions to: display, via an interface, an output of a portion of a video recommended for preview to facilitate limiting the portion of the video for one or more audience groups during playback, receive, via the interface and for a user profile, an indication of whether to limit the portion of the video for one or more audience groups during playback, and store the indication for limiting use in the portion of the video for one or more audience groups during playback.

[0005] In another example, a computer readable device is provided having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations to restrict video content. The operations include: displaying, via an interface, an output of a portion of a video recommended for preview to facilitate restricting the portion of the video for one or more audience groups during playback, receiving, via the interface and for a user profile, an indication of whether to restrict the portion of the video for one or more audience groups during playback, and storing the indication for use in restricting the portion of the video for one or more audience groups during playback.

[0006] To achieve the above and related purposes, one or more aspects include features fully described below and particularly pointed out in the claims. The following description and the accompanying drawings set forth in detail the illustrative features of one or more aspects. However, these features are only a few of the various ways in which the principles of the various aspects can be employed, and this description is intended to include all such aspects and their equivalents. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 is a schematic diagram of an example of an apparatus for limiting video content according to aspects described herein.

[0008] Figure 2 is a flow chart of an example of a method for limiting video content according to aspects described herein.

[0009] Figure 3 is a flow diagram of an example of a method for receiving an indication of a portion of a video to be previewed for restriction according to aspects described herein.

[0010] Figure 4

[0013] Illustrated is an example of a service architecture for providing video content restrictions according to aspects described herein.

[0011] Figure 5 is a schematic diagram of an example of a service architecture for detecting restricted content via a conversation, using speech-to-text processing or topic modeling according to aspects described herein.

[0012] Figure 6 is a schematic diagram of an example of an apparatus for performing the functions described herein according to aspects described herein. DETAILED DESCRIPTION

[0013] The specific embodiments provided below in conjunction with the accompanying drawings are intended to illustrate various configurations and are not intended to represent the only configuration in which the concepts described herein can be practiced. The specific embodiments include specific details in order to provide a thorough understanding of the various concepts. However, it is obvious to those skilled in the art that these concepts can be practiced without these specific details. In some cases, in order to avoid blurring these concepts, well-known components are shown in block diagram form.

[0014] The present disclosure describes various examples of facilitating restrictions on video content by providing portions of video content for preview and indicating whether to restrict portions of the video content when playing the video content for one or more audience groups. For example, for given video content (such as a movie, television program, etc.), a user interface can be provided to identify, represent, or otherwise display portions of the video content that are recommended for review for possible restriction. The user interface can facilitate the restriction of portions of the video content based on receiving input indicating that portions of the content for one or more audience groups should be restricted.

[0015] In an example, portions of video content to be recommended for review may be selected using one or more of a variety of mechanisms, including content-based mechanisms, such as artificial intelligence (AI)-based mechanisms that may identify portions of video content based on providing natural language or speech-to-text processing of closed captions or dialogue as input to a machine learning (ML) model trained on sensitive phrases, or providing images from scenes or frames of a video as input to an ML model trained on known sensitive images. In another example, a ML model-based recommendation system using contextual filtering or collaborative filtering may be trained based on natural language or speech-to-text processing of closed captions, or image content of other scenes identified by similar users as restricted (e.g., for a given audience group). In another example, the mechanism for selecting portions of the video content may include a contextual collaborative filtering based recommendation system that may select portions of the video content based on portions of the video content that other users (e.g., users with similar profiles to the user to whom the preview is being provided, or vice versa) have indicated are to be restricted, or portions of the video content based on portions that are similar to other portions of the video content that other users (e.g., users based on similar profiles to the user to whom the preview is being provided, or vice versa) have identified as being restricted, etc. In this regard, custom video restrictions may be provided using custom segmentation of the video and recommendations for portions of the video to be previewed, where the recommendations may be AI based or other content based, collaborative filtering based, etc.

[0016] AI-based models can provide recommendations about safe video content boundaries with preview options so that users / viewers can quickly decide to lock safe content boundaries without having to manually watch each scene to mark safe points. This intelligence can provide more recommendations by analyzing restriction decisions or other configurations made by other users, thereby extending it to collaborative filtering based models to make it more efficient and accurate. In some examples described in this article, the model can become a fusion of automatic intelligence combined with human intelligence to make recommendations in a more accurate manner. This model or design can help parents control any video and transform it into safe content suitable for their family, ensuring the safety of children and / or creating positive influence through safe content videos.

[0017] In one example, a user may select the type of mechanism to be used when generating recommendations for preview, such as an AI-based mechanism, a collaborative filtering-based mechanism, a combination thereof, and / or the like. In this regard, presenting recommendations for preview may allow a user to more easily restrict content in a video for certain audience groups associated with the user (e.g., children of a certain age), and restricted content may be removed from the video during playback. In these examples, a user may not need to watch the entire video content to identify material to be restricted, but may only need to view and suggest recommended portions of the video content, which may save time and improve the user experience when restricting video content.

[0018] According to aspects described herein, restricted content can be detected via AI / image review models and natural language processing (NLP)-based speech-to-text processing of video content to indicate the possibility of sensitive content in a particular frame. In some aspects, in addition to automatic detection, recommendations for sensitive content can be provided using collaborative filtering-based recommendations from similar user profiles and content-based recommendations using user profile history. In some aspects, restricted frames collected through automatic detection of restricted content and collaborative filtering-based recommendations can be displayed to the user as a preview via a user interface, and / or the user interface can allow the user to indicate whether to restrict the content, as further described herein.

[0019] Now go to Figure 1-Figure 6 , the figure depicts examples involving one or more components and one or more methods that can perform the actions or operations described herein, wherein the components and / or actions / operations represented by dashed lines may be optional. Figure 2 , Figure 3 and Figure 5The operations described in the description are presented in a particular order and / or performed by example components, but in some examples, the order of actions and the components performing the actions may vary depending on the implementation. In addition, in some examples, one or more actions, functions, and / or components described may be performed by a specially programmed processor, a processor executing specially programmed software or computer-readable media, or by any other combination of hardware components and / or software components capable of performing the described actions or functions.

[0020] Figure 1 1 is a schematic diagram of an example of a device 100 (e.g., a computing device) for performing functions related to limiting video content. In an example, the device 100 may include a processor 102 and / or a memory 104 configured to execute or store instructions or other parameters related to providing an operating system 106, which may execute one or more applications or processes, such as, but not limited to, a video restriction component 110 for defining one or more video restrictions for one or more user profiles or audience groups within a plurality of user profiles, and / or a video playback component 112 for playing a video based on one or more defined video restrictions for a given user profile or audience group. For example, the processor 102 and the memory 104 may be separate components communicatively coupled via a bus (e.g., on a motherboard or other portion of a computing device, on an integrated circuit (e.g., a system-on-chip (SoC)), etc.), components integrated with each other (e.g., the processor 102 may include the memory 104 as an onboard component), and / or the like. The memory 104 may store instructions, parameters, data structures, etc. for use / execution by the processor 102 to perform the functions described herein.

[0021] For example, the video restriction component 110 may include a preview component 114 that may display an output of a portion of the video for preview, and / or a restriction indication component 116 that may receive input of whether to restrict a portion of the video content for a user profile, one or more audience groups associated with the user profile, etc., which may be determined based on the display of the preview. In an example, the video restriction component 110 may receive the portion of the video for preview, and / or may store the restriction indication via the memory 104 local to the device 100 and / or via other components on the network 120. In fact, almost all of the components described herein may be located within the device 100, or in other devices or nodes at distributed locations accessible via the network 120.

[0022] In an example, a portion acquisition component 122 may be provided (e.g., in the device 100 or in another node accessible via the network 120) for acquiring portions of a video for preview by the video restriction component 110. In various examples described herein, the portion acquisition component 122 may utilize one or more machine learning (ML) models 124 to determine which portions of a video are to be presented for preview, so that a user can filter or restrict content for certain audience groups without having to watch the entire video. In addition, for example, a video storage component 126 may be provided (e.g., in the device 100 or in another node accessible via the network 120) for storing videos accessed and played by the video playback component 112. For example, the video storage component 126 may be operated by a video streaming service, and the video playback component 112 of the device 100 (and / or other video playback components 132 of other devices 130) may receive the video streaming service to stream videos from the video storage component 126.

[0023] In addition, for example, a profile / restriction storage component 128 may be provided (e.g., in the device 100 or in another node accessible via the network 120) for storing user profile information (e.g., profile demographics, household user information, or audience groups associated with the user profile, etc.) and / or restrictions indicated for the user profile or the associated audience groups. In an example, the video playback component 112 or the video playback component 132 of another device 130 may communicate with the profile / restriction storage component 128 based on the associated user profile to determine restrictions for playing video content stored in the video storage component 126 for the user profile and / or for one or more associated audience groups.

[0024] In some examples, a streaming video subscription provider storing videos in a video storage component 126 may upload content to a portion acquisition component 122, which may be part of an AI-based product recommendation model that may process content based on a rating policy and / or apply image processing algorithms to segment the video to detect portions (e.g., scenes, frames, etc.) that include different types of sensitive content (e.g., violence / adult content, adult language, smoking, etc.), and the output of such processed content boundaries may be persisted in a cloud storage device. In an example, the output of the AI ​​model may be fused with other restrictions indicated in a user profile and stored in a profile / restriction storage component 128, which may be part of the same product, to generate a fused recommendation. At the time of customization, these two options may be presented to the user via the video restriction component 110.

[0025] In the user interface, through the video restriction component 110, the user may have the option to select AI-based recommendations (either full AI or AI plus hybrid (artificial intelligence based on other user profiles)), which may include a manual customization function to set the boundaries of the content range according to the user's preferences. The recommendation options may be provided to the portion acquisition component 122, which in turn affects how the portion acquisition component 122 acquires and provides portions of the video content to the video restriction component 110 (e.g., using AI-enabled recommendations based on one or more ML models 124, and / or providing recommendations based on restriction indications of other user profiles stored by the profile / restriction storage component 128).

[0026] This solution can also be used in enterprise / business services to limit the content of training videos in any enterprise / business service, especially when some content may not be relevant to certain regions. The part acquisition component 122 can acquire these parts for review, thereby allowing the indication of restrictions based on the region and policy. For example, in the business field, content related to "invoice whitelist" should not be included in the training video, because the whitelist refers to a category of a specific group of people. Therefore, any such content can be presented for preview and possible indication of restrictions (such as profiles associated with engineers living in the United States).

[0027] Furthermore, in some examples, in addition to families, schools and institutions can also benefit from this model because they can customize the video content to align with their own institutional culture and fundamental principles, thus becoming safe content. The producers of the original video cannot customize the content for each user because they cannot measure each person's needs, including cultural values.

[0028] Figure 2 is a flow chart of an example of a method 200 for limiting video content. For example, the method 200 may be performed by the device 100 and / or one or more components thereof to facilitate displaying a portion of a previous video and / or facilitating limiting the same.

[0029] In method 200, at action 202, an output of a portion of a video may be displayed, the portion of the video content being recommended for preview to facilitate limiting the portion of the video content for one or more audience groups during playback. In an example, a preview component 114 in a video restriction component 110 (e.g., in combination with a processor 102, a memory 104, an operating system 106, etc.) may display an output of a portion of the video recommended for preview to facilitate limiting the portion of the video for one or more audience groups during playback. For example, the preview component 114 may display the portion of the video on a user interface, wherein the user interface may also allow portions of the video to be marked as being restricted for one or more audience groups associated with a user profile. In this regard, the video restriction component 110 may allow a user to preview a limited portion of a video without having to watch the entire video to set restrictions.

[0030] In method 200, optionally, at act 204, an indication of a portion of a video may be received. For example, a preview component 114 in video restriction component 110 (e.g., in conjunction with processor 102, memory 104, operating system 106, etc.) may receive (e.g., from portion acquisition component 122) an indication of a portion of a video. For example, for a given video to be streamed from video storage component 126, portion acquisition component 122 may provide video restriction component 110 with a list of portions of the video to be previewed for possible content restriction. For example, the list of portions may include a start and stop time for each portion. In any case, preview component 114 may display (or allow the user to select a portion) for each portion in the list of portions, along with the option to preview the portion and indicate whether the portion is to be restricted for one or more audience groups associated with a user profile. Figure 3 Various options for receiving indications are displayed.

[0031] Figure 3 A flow chart illustrating an example of an act or operation of generating an indication of a portion of a received video. At act 204, receiving the indication may include or be based on Figure 3 In an example, which actions to include can be based on a selection made via an interface of the video restriction component 110, the selection indicating which type of recommendation is requested, as further described herein. In an example, the portion acquisition component 122 can apply the actions to acquire portions of the video, and the video restriction component 110 can receive indications of the portions of the video from the portion acquisition component 122 accordingly. For example, as described above and further herein, the recommendation can be contextually based or based on collaborative filtering.

[0032] For example, upon receiving an indication of a portion of a video at act 204 (optionally at act 302), a vector of natural language from subtitles or speech-to-text processing of the video may be input into the ML model to identify the portion of the video. In an example, the portion acquisition component 122 (e.g., based on a request from the video restriction component 110, and / or in conjunction with the processor 102, memory 104, operating system 106, etc. of a node operating the portion acquisition component 122) may input a vector of natural language from subtitles or speech-to-text processing of the video into the ML model 124 to identify the portion of the video. For example, the portion acquisition component 122 may acquire subtitle text associated with the video content, or may perform speech-to-text processing (or receive the result of the processing) on ​​the dialogue of the video content (e.g., the audio track of the video), and provide the natural language as input to the ML model 124. The ML model 124 may, for example, be trained on natural language determined to include sensitive material to provide context-based recommendations. For any portion of the video content determined to include sensitive material based on the output of the ML model 124 given a natural language input, the portion acquisition component 122 can determine the start and stop times or other identifying information for the portion of the video and can provide the relevant information in an indication to the video restriction component 110.

[0033] In an example, the ML model 124 can be trained using unsupervised topic modeling, which can eliminate the need for domain knowledge expertise when identifying scenes. For example, applying topic modeling to a set of frames can give a number (n) of topics that can indicate scenes with restricted / sensitive content. The topic model can be mapped to an existing dictionary that includes offensive words or other words that indicate restricted / sensitive content. If a topic in a set of frames maps to restricted / sensitive content, the start and stop times of the frame can be saved and the frame can be skipped during subsequent playback for a specific audience group.

[0034] Figure 5 An example of a service architecture 500 for detecting restricted content using speech-to-text processing or topic modeling via dialogue according to aspects described herein is illustrated. According to the service architecture 500, dialogue can be utilized to detect restricted content in a frame, for example. One or more movies can be obtained at 502. If it is determined at 504 that the movie has subtitles (e.g., in metadata received with the movie at 502), the movie subtitles can be obtained for each frame at 506. If there are no subtitles, a speech-to-text application or application programming interface (API) can be used at 508 to obtain or otherwise determine the subtitles. At 510, the subtitles or dialogue for each frame can be text-processed, which can include using natural language processing techniques, such as removing stop words, stemming and / or lemmatization, and part-of-speech tagging.

[0035] After text processing / preprocessing, topic modeling can be performed at 512 to map topic models to terms in an interpersonally offensive language dictionary at 514, and / or to possibly detect the occurrence of offensive language in the conversation. In an example, topic modeling can be performed on each frame to obtain the most numerous topic models (e.g., the top 5 topic models) at 512. If any topic model maps to an interpersonally offensive word or phrase, the frame can be marked as restricted, and the start and stop times of the frame can be saved for use when playing the movie, as described herein (e.g., to skip or otherwise remove restricted content during playback).

[0036] Reference again Figure 3 In another example, upon receiving an indication of a portion of a video at act 204 (optionally at act 304), one or more images from the video may be input into an ML model to identify the portion of the video. In an example, the portion acquisition component 122 (e.g., based on a request from the video restriction component 110, and / or in conjunction with a processor 102, a memory 104, an operating system 106, etc. of a node operating the portion acquisition component 122) may input one or more images from the video into an ML model 124 to identify the portion of the video. For example, the portion acquisition component 122 may provide an image from a video stream as input to the ML model 124. The ML model 124 may, for example, be trained on images determined to include sensitive content. For any portion of the video determined to include sensitive content based on the output of the machine learning model 124 given an image input, the portion acquisition component 122 may determine the start and stop times or other identifying information of the portion of the video, and may provide the relevant information to the video restriction component 110 in the indication.

[0037] In another example, upon receiving an indication of a portion of the video at act 204 (optionally at act 306), a portion of the video may be identified as being similar to another portion of another video indicated as restricted in the user profile. In an example, the portion acquisition component 122 (e.g., based on a request from the video restriction component 110, and / or in conjunction with the processor 102, memory 104, operating system 106, etc. of a node operating the portion acquisition component 122) may identify that the portion of the video is similar to another portion of another video indicated as restricted in the user profile. In an example, the portion acquisition component 122 may input natural text or images of other video content indicated as restricted in the user profile into the ML model 124 (e.g., when training the ML model 124), so that the ML model 124 may output portions of similar videos. In an example, the ML model 124 may be specific to a user profile, or may be specific to certain user profile parameter values ​​(e.g., similar parameter values ​​for one or more parameters such as demographic or cultural parameters, family member parameters, country, region, language preference, gender, etc.), such that user profiles with similar parameter values ​​may use the same ML model 124 to identify portions of possible videos for preview as a recommendation based on collaborative filtering. In another example, the portion acquisition component 122 may acquire natural text of portions of other video content that the user profile indicates is restricted for one or more audience groups, and may detect the same or similar natural text in portions of the video to provide portions of the video for preview. For example, the portion acquisition component 122 may acquire natural text of portions of other video content that is restricted from the profile / restriction storage component 128.

[0038] In another example, after receiving an indication of a portion of the video at act 204 (optionally at act 308), the portion of the video may be identified as being restricted by another user profile having one or more parameter values ​​similar to those in the user profile. For example, the portion acquisition component 122 may identify the portion of the video as being restricted by another user profile having one or more parameter values ​​similar to (or the same as) those in the user profile based on a request from the video restriction component 110, and / or in conjunction with the processor 102, memory 104, operating system 106, etc. of a node operating the portion acquisition component 122. In an example, the portion acquisition component 122 may input natural text or images of the portion of the video restricted by the other user profile into the ML model 124 (e.g., when training the ML model 124), so that the ML model 124 may output a similar portion of the video. In an example, the ML model 124 may be specific to a user profile, or may be specific to certain user profile parameter values ​​(e.g., similar parameter values ​​for one or more parameters such as demographic or cultural parameters, family member parameters, country, region, language preference, gender, etc.), such that user profiles with similar parameter values ​​may use the same machine learning model 124 to identify portions of possible videos for preview as a collaborative filtering-based recommendation as described.

[0039] In another example, the portion acquisition component 122 can obtain natural text of portions of other videos indicated by other user profiles as restricted for one or more audience groups, and can detect the same or similar natural text in portions of the video to provide portions of the video for preview to the user profile. In this regard, for example, the portion acquisition component 122 can use content-based collaborative filtering to identify portions of the video and determine them as videos restricted by another user profile. For example, the portion acquisition component 122 can obtain natural text of portions of other videos indicated by other user profiles as restricted from the profile / restriction storage component 128. In an example, the portion acquisition component 122 can obtain this information based on inputting the user profile and the other user profiles into the ML model 124 to obtain a similarity score or confidence, etc. to determine that certain parameter values ​​in the user profile and the other user profiles are similar.

[0040] In another example, after receiving an indication of a portion of the video at act 204 (optionally at act 310), the portion of the video may be identified as being similar to another portion of another video restricted by another user profile having one or more parameter values ​​similar (or the same) as the user profile. For example, the portion acquisition component 122 may identify the portion of the video as being similar to another portion of another video restricted by another user profile having one or more parameter values ​​similar (or the same) as the user profile based on a request from the video restriction component 110 and / or in conjunction with the processor 102, memory 104, operating system 106, etc. of a node operating the portion acquisition component 122. In an example, the portion acquisition component 122 may input natural text or images of the portion of the video of the other video restricted by the other user profile into the ML model 124 (e.g., when training the ML model 124), so that the ML model 124 may output the portion of the video that is similar to the other portion of the other video. In an example, the ML model 124 can be specific to a user profile, or can be specific to certain user profile parameter values ​​(e.g., similar parameter values ​​for one or more parameters such as demographic or cultural parameters, family member parameters, country, region, language preference, gender, etc.), such that user profiles with similar parameter values ​​can use the same ML model 124 to identify portions of possible videos for preview as collaborative filtering-based recommendations as described. In an example, the portion acquisition component 122 can acquire this information based on determining that certain parameter values ​​in the user profile and other user profiles are similar by inputting the user profile and other user profiles into the ML model 124 to obtain a similarity score or confidence, etc.

[0041] Return to Reference Figure 2 In method 200, at act 206, an indication may be received for a user profile indicating whether a portion of a video is to be restricted for one or more audience groups during playback. In an example, a restriction indication component 116 of a video restriction component 110, for example in combination with a processor 102, a memory 104, an operating system 106, etc., may receive an indication for a user profile indicating whether a portion of a video is to be restricted for one or more audience groups during playback. For example, the video restriction component 110 may provide an option via a user interface to indicate that a portion of a video is to be restricted for one or more audience groups. In an example, a user profile may be associated with different audience groups, which may be specified in a profile / restriction storage component 128. For example, a user profile may specify an audience group for each of a plurality of family members of the user profile. For example, an audience group may be limited based on age or age range, content preferences, and the like.

[0042] In an example, when previewing content, the video restriction component 110 can provide an option to indicate whether to limit content for each of the multiple audience groups associated with the user profile. In this regard, for example, a portion can be limited for one audience group, but not for another audience group. In a specific example, the video restriction component 110 can provide a first option to limit the content of the 8-year-old and below audience group (which can correspond to a child in the family), and provide a second option to limit the content of the 9-12-year-old audience group (which can correspond to another child in the family). Therefore, for example, a user can indicate that the portion of the video of the 8-year-old and below audience group is limited, but the portion of the video of the 9-12-year-old audience group is not limited. In any case, the restriction indication can be stored for the video (for example, by the profile / restriction storage component 128), and when playing a video for a given audience group associated with the user profile, it can be accessed by the video playback component 112 or 132.

[0043] In method 200, at act 208, instructions may be stored for limiting portions of a video for one or more audience groups during playback of the video. For example, a restriction indication component 116 in video restriction component 110 (e.g., in conjunction with processor 102, memory 104, operating system 106, etc.) may store (e.g., via profile / restriction storage component 128) instructions for limiting portions of a video for one or more audience groups during playback. For example, restriction indication component 116 may store instructions and one or more parameters indicating portions of a video (e.g., start time and stop time or other identifying information within the video). For example, restriction indication component 116 may store instructions in user profile information and / or information related to one or more audience groups as described.

[0044] In method 200, optionally at act 210, an audience group of the one or more audience groups requesting to play the video may be identified. In an example, video playback component 112 or 132 (e.g., in combination with processor 102, memory 104, operating system 106, etc. (or similar components of device 130)) may identify an audience group of the one or more audience groups requesting to play the video. For example, video playback component 112 or 132 may receive a request to access a video from a video streaming service (e.g., from video storage component 126), and the request may be associated with a user profile and / or a corresponding audience group.

[0045] In method 200, optionally at action 212, stored indications may be retrieved. In an example, video playback component 112 or 132 (e.g., in combination with processor 102, memory 104, operating system 106, etc. (or a similar component of device 130)) may retrieve stored indications. For example, video playback component 112 or 132 may retrieve an indication to restrict a portion of a video from profile / restriction storage component 128, e.g., along with indications to restrict other portions of other videos. For example, video playback component 112 or 132 may retrieve a user profile associated with a video playback request and retrieve restriction indications for the user profile, associated with the video, and / or associated with a particular audience group.

[0046] In method 200, optionally at act 214, playback of the video with the portion of the video removed based on the indication is displayed for the audience group and based on the request, or playback of the video without the portion of the video removed based on the indication is displayed. For example, video playback component 112 or 132 (e.g., in combination with processor 102, memory 104, operating system 106, etc. (or similar components of device 130)) can display playback of the portion of the video removed or not removed based on the indication for the audience group based on the request. For example, video playback component 112 or 132 can obtain a list of restriction indications requested for playback by the audience group as described, and remove or skip the portion of the video indicated as restricted. For example, video playback component 112 or 132 can determine a start time and an end time for each restriction indication and remove the portion of the video from the playback stream during playback, or skip starting at the start time and continuing at the end time, etc.

[0047] In an example, as described, content restrictions may be subscription-based. Accordingly, in one example of method 200, optionally in action 216, a subscription of a user profile may be verified as having subscription capabilities for content restrictions based on audience groups. In an example, video playback component 112 or 132 (e.g., in combination with processor 102, memory 104, operating system 106, etc. (or similar components of device 130)) may verify the user profile as having subscription capabilities for content restrictions based on audience groups before obtaining restriction indications or otherwise restricting content. In an example, subscription capabilities may be indicated based on audience groups, such that a user may have different subscription levels for restricting one audience group (or specific audience groups), restricting more than one audience group (or different specific audience groups), and so on.

[0048] In other examples, as described, the restriction indication can be used to train one or more ML models 124. In the method 200, optionally at act 218, the ML model can be trained based on the indication and / or one or more parameters of the user profile. In an example, the profile / restriction storage component 128 (e.g., in combination with the processor 102, memory 104, operating system 106, etc. of a node including the profile / restriction storage component 128) can train the ML model (e.g., one or more ML models of the plurality of ML models 124) based on the indication and / or one or more parameters of the user profile. For example, the profile / restriction storage component 128 can train the ML model 124 based on the portion marked as a restricted portion, the attributes of the portion of the video (e.g., natural language in the audio or closed captions of the portion of the video, the image of the portion of the video, etc.), the user profile characteristics (e.g., demographic parameters, cultural parameters, family member parameters, etc.). In an example, as described, the portion acquisition component 122 can use the trained ML model 124 to recommend portions of a video (e.g., portions of the video used to train the ML model 124 or similar portions of the video or other videos) for preview for certain user profiles or associated audience groups. In addition, for example, as described, the profile / restriction storage component 128 can train the ML model 124 based on natural language using K-means of probabilities of keywords generated for each topic model.

[0049] Figure 4 An example of a service architecture 400 providing video content restrictions according to aspects described herein is illustrated. Service architecture 400 may include streaming service 402, which may provide streaming video content to different users based on user profiles or subscriptions with streaming service. Streaming service may communicate with content restriction service 404 to obtain information on restricting content during video streaming for a given user profile. For example, a user profile of streaming service 402 may be associated with a user profile in content restriction service 404. In an example, a user may subscribe to streaming service 402 (and / or multiple streaming services), and separately subscribe to content restriction service 404, wherein content restriction service 404 may interact with streaming service 402 (and / or multiple streaming services), to provide restriction information that may be used when streaming service 402 (and / or multiple streaming services) streams video content to a user.

[0050] In an example, the streaming service can access the video content via the video storage component 126 for playback. As described, the content restriction service 404 can also access the video via the video storage component 126 for content restriction. In addition, as described, the content restriction service 404 can be provided by the device 100, or by one or more components distributed across multiple devices, which can include the video restriction component 110 and / or its components (e.g., the preview component 114, the restriction indication component 116, etc.), the portion acquisition component 122, the ML model 124, the profile / restriction storage component 128, etc. The content restriction service 404 can access these components to facilitate the definition of restrictions on portions of the video for certain user profiles or audience groups, the selection of portions of the video to be previewed for indicating restrictions, the enforcement of the indicated restrictions when streaming the video content, and the like.

[0051] In an example, as described, the content restriction service 404 may include a process for previewing AI and / or custom recommendations 406 for portions of a video that may include sensitive material. As described, AI-based recommendations may be provided by ML models, such as n AI / image censorship models 408, which may take as input a vector of natural language from closed captions, or a vector of natural language from speech-to-text processing of an audio track of a video, or a vector of an image from a video, and may indicate one or more portions of the video for preview. In addition, as described, custom recommendations may be provided based on a user configuration 410 or other user profile of a user in the content restriction service, such as similar restrictions provided by other user profiles for the video, similar restrictions provided by the user or other user profiles for similar portions of other videos, and the like. In an example, a user may enable AI-based recommendations, custom recommendations, additional recommended types, and / or combinations thereof for preview. In this regard, for example, the content restriction service 404 may include a process for configuring AI and custom recommendations 412, which may provide an option on a user interface for a user to configure the type of recommendation.

[0052] For example, the content restriction service 404 may include a process for content restrictions 414 that allows a user to indicate content restrictions on portions of a video for one or more audience groups that may be associated with a user profile. The content restriction service 404 may store the content restrictions 414 as part of the user profile, or metadata associated with the user profile, and the content restrictions 414 may be specific to a given audience group. In any case, for example, the content restriction service 404 may provide content restriction information to the streaming service 402 for a given user profile or associated audience group to allow the streaming service 402 to restrict video streaming content based on the restrictions. In other examples, as described herein, the content restriction service 404 may also use the content restrictions 414 to generate other recommendations for content restrictions (e.g., for the same user profile or a different user profile), to train an ML model that provides AI-based content restriction recommendations, to include other recommendations in a user configuration for custom recommendations, and so on.

[0053] Figure 6 An example of a device 600 is shown, including Figure 1 . In one aspect, device 600 may include a processor 602, which may be similar to processor 102, for performing processing functions associated with one or more components and functions described herein. Processor 602 may include a single or a collection of multiple processors, or a multi-core processor. In addition, processor 602 may be implemented as an integrated processing system and / or a distributed processing system.

[0054] The device 600 may also include a memory 604, which may be similar to the memory 104, for use, for example, in storing a local version of an operating system (or components thereof) and / or applications executed by the processor 602, such as the video restriction component 110, the video playback component 112, the portion acquisition component 122, the ML model(s) 124, the video storage component 126, the profile / restriction storage component 128, etc. The memory 604 may include any type of computer-usable memory, such as random access memory (RAM), read-only memory (ROM), tape, magnetic disk, optical disk, volatile memory, non-volatile memory, and any combination thereof.

[0055] Additionally, device 600 may include a communication component 606 that utilizes the hardware, software, and services described herein for establishing and maintaining communications with one or more other devices, parties, entities, and the like. Communication component 606 may facilitate communications between components on device 600, as well as communications between device 600 and external devices, such as devices located on a communication network and / or devices serially or locally connected to device 600. For example, communication component 606 may include one or more buses, and may also include transmit chain components and receive chain components associated with wireless or wired transmitters and receivers, respectively, which are operable to interface with external devices.

[0056] In addition, the device 600 may also include a data store 608, which may be any suitable combination of hardware and / or software for providing mass storage of information, databases, and programs related to aspects described herein. For example, the data store 608 may be or may include a data repository for an operating system (or components thereof), an application, related parameters (not currently executed by the processor 602), etc. In addition, the data store 608 may also be a data repository for the video restriction component 110, the video playback component 112, the partial acquisition component 122, the (multiple) ML model 124, the video storage component 126, the profile / restriction storage component 128, and / or one or more other components of the device 600.

[0057] The device 600 may also optionally include a user interface component 610 that is capable of receiving input from a user of the device 600 and further operable to generate output for presentation to the user. The user interface component 610 may include one or more input devices, including but not limited to a keyboard, a numeric keypad, a mouse, a touch screen display, navigation keys, function keys, a microphone, a voice recognition component, a gesture recognition component, a depth sensor, a gaze tracking sensor, a switch / button, any other mechanism capable of receiving input from a user, or any combination thereof. In addition, the user interface component 610 may also include one or more output devices, including but not limited to a display, a speaker, a tactile feedback mechanism, a printer, any other mechanism capable of presenting output to a user, or any combination thereof.

[0058] By way of example, an element, any part of an element, or any combination of elements can be implemented using a "processing system" comprising one or more processors. Examples of processors include microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other suitable hardware configured to perform the various functions described in this disclosure. One or more processors in a processing system can execute software. Software should be broadly interpreted as instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, processes, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description languages, or other.

[0059] Therefore, in one or more aspects, one or more functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software form, these functions may be stored or encoded as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media. Storage media can be any available media that can be accessed by a computer. By way of example, but not limitation, such computer-readable media may include random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store the required program code in the form of instructions or data structures and can be accessed by a computer. Disks and optical disks used in this article include compact disks (CDs), laser disks, optical disks, digital versatile disks (DVDs), and floppy disks, where disks typically reproduce data magnetically, while optical disks reproduce data optically by lasers. The above combinations should also be included in the scope of computer-readable media.

[0060] The foregoing description is intended to enable those skilled in the art to practice the various aspects described herein. Those skilled in the art will readily appreciate the various modifications to these aspects, and the general principles defined herein may be applied to other aspects. Therefore, the claims are not intended to be limited to the aspects shown herein, but to be consistent with the full scope of the claim language, wherein, unless otherwise specified, references to singular elements are not intended to mean "one and only one", but rather "one or more". Unless otherwise specifically stated, the term "some" refers to one or more. All structural and functional equivalents of the elements of the various aspects described herein, whether known to those skilled in the art or later, are expressly included and intended to be covered by the claims. In addition, regardless of whether such disclosure is explicitly stated in the claims, nothing disclosed herein is intended to be dedicated to the public. Unless the element is explicitly stated using the phrase "device for...", no claim element shall be interpreted as a "device plus function" form.

Claims

1. A computer-implemented method for limiting video content, comprising: receiving a recommendation of a portion of a video to be displayed for review to facilitate potentially limiting the portion of the video for one or more audience groups during playback, wherein receiving the recommendation comprises receiving at least a first recommendation of a first portion of the video as a context-based recommendation, and receiving a second recommendation of a second portion of the video as a collaborative-based recommendation; displaying, via an interface, an output of the portion of the video recommended for preview to facilitate potentially limiting the portion of the video for the one or more audience groups during playback; receiving, via the interface and for a user profile, an indication of whether to restrict the portion of the video for the one or more audience groups during playback; as well as The indication is stored for use in limiting usage of the portion of the video for the one or more audience groups during playback.

2. The computer-implemented method of claim 1 , further comprising: identifying an audience group among the one or more audience groups that requested playing of the video; Based on requesting playback of the video, obtaining the stored indication; as well as For the audience group and based on the request, a playback of the video is displayed with the portion of the video removed based on the indication, or a playback of the video is displayed without the portion of the video removed based on the indication.

3. The computer-implemented method of claim 2, wherein displaying the playback of the video for the audience group is based on the ability to verify the subscription of the user profile as a subscription with content restrictions based on the audience group.

4. The computer-implemented method of claim 1 further comprises receiving the context-based recommendation into a machine learning (ML) model based on at least one input of a vector of natural language from closed captioning or speech-to-text processing of the video, or a vector of an image from the video, to identify the first portion of the video as including restricted content.

5. The computer-implemented method of claim 4 further comprises training the machine learning model based on indications from a plurality of indications corresponding to the first portion of the video having restricted content and one or more parameters of the user profile.

6. The computer-implemented method of claim 1, further comprising receiving the second recommendation based on collaborative filtering to identify the second portion of the video as being restricted by another user profile having one or more parameters having similar values ​​to the user profile.

7. The computer-implemented method of claim 1, further comprising receiving the second recommendation based on collaborative filtering to identify the second portion of the video as similar to another portion of another video indicated as restricted in the user profile.

8. A computer-implemented method according to claim 7, wherein identifying the second portion of the video as similar to the other portion of the other video is based on: inputting a vector associated with the second portion of the video into a machine learning (ML) model, and receiving an output of the other portion of the other video, and identifying that the user profile indicates that the other portion of the other video is restricted.

9. The computer-implemented method of claim 8 further comprises training the machine learning model based on indications from the plurality of indications corresponding to the second portion of the video and one or more parameters of the user profile.

10. The computer-implemented method of claim 1, wherein the user profile defines the one or more audience groups to include one or more family members based on the age of the one or more family members.

11. A computer-implemented method according to claim 1, wherein displaying the output provides a restricted preview based on a selection of a type of recommendation including one or more of the following: context-based recommendations, collaborative-based recommendations with restrictions indicated by similar user profiles, or collaborative-based recommendations of the user profile for other videos.

12. A device for limiting video content, comprising: a memory for storing instructions; as well as at least one processor coupled to the memory and configured to execute instructions for: receiving a recommendation of a portion of a video to be displayed for review to facilitate potentially limiting the portion of the video for one or more audience groups during playback, wherein receiving the recommendation comprises receiving at least a first recommendation of a first portion of the video as a context-based recommendation, and receiving a second recommendation of a second portion of the video as a collaborative-based recommendation; displaying, via an interface, an output of the portion of the video recommended for preview to facilitate potentially limiting the portion of the video for the one or more audience groups during playback; receiving, via the interface and for a user profile, an indication of whether to restrict the portion of the video for the one or more audience groups during playback; as well as The indication is stored for use in limiting usage of the portion of the video for the one or more audience groups during playback.

13. The apparatus of claim 12, wherein the at least one processor is configured to execute instructions for: identifying an audience group among the one or more audience groups that requested playing of the video; Based on requesting playback of the video, obtaining the stored indication; as well as For the audience group and based on the request, a playback of the video is displayed with the portion of the video removed based on the indication, or a playback of the video is displayed without the portion of the video removed based on the indication.

14. The apparatus of claim 13, wherein the at least one processor is configured to execute instructions for: displaying playback of the video for the audience group based on verifying the subscription of the user profile as a subscription with content restrictions based on the audience group.

15. A computer-readable device having instructions stored thereon, which, when executed by at least one computing device, cause the at least one computing device to perform the operations of the computer-implemented method according to any one of claims 1-11.

Images