Identification and tracking of Internet memes
By monitoring and comparing multiple sources of digital visual content, identifying and tracking Internet memes using perceptual hash algorithms, the problem of difficult measurement of virality in the prior art is solved, and accurate impact measurement and tracking of Internet memes is achieved.
Patent Information
- Application Number
- CN202311490310.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2020-02-24
- Filing Date
- 2021-02-23
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-02-23
Smart Images

Figure CN117456209B_ABST
Abstract
Description
[0001] This application is a divisional application of the Chinese invention patent application with application number 202180030631.6, application date February 23, 2021, and invention name “Identification and Tracking of Internet Memes”.
[0002] Priority claim
[0003] This application claims priority to U.S. patent application Ser. No. 16 / 799,641, filed on Feb. 24, 2020, the entire contents of which are incorporated herein by reference. Background Art
[0004] The dramatic increase in Internet access has significantly expanded participation in social networks and social media platforms as a means of sharing information. Internet memes are examples of information shared over the Internet (typically using social networks and social media platforms). Internet memes are fragments of digital media (e.g., digital images or digital video clips) that are combined with text words or phrases to convey or represent a specific theme or meaning. Internet memes can be used to promote or share different types of information including humor, news, entertainment, etc. The goal of many Internet meme creators and promoters is to make an Internet meme go "viral," where the sharing of a viral Internet meme surges rapidly in a short period of time. BRIEF DESCRIPTION OF THE DRAWINGS
[0005] The accompanying drawings are incorporated into and form a part of the detailed description, and the accompanying drawings show various embodiments of the present subject matter and are used together with the detailed description to illustrate the principles of the subject matter discussed below. Unless explicitly noted, the drawings mentioned in the description of the drawings should be understood as not being drawn to scale. In this article, similar items are marked with similar item numbers.
[0006] Figure 1A An example computer system is shown upon which the embodiments described herein may be implemented.
[0007] Figure 1B An example communication network is shown upon which embodiments described herein may be implemented.
[0008] Figure 2 is a block diagram illustrating a system for identifying and tracking Internet memes, according to an embodiment.
[0009] Figure 3 is a block diagram illustrating an Internet meme determiner of a system for identifying Internet memes according to an embodiment.
[0010] Figure 4A and Figure 4B Shown are examples of different iterations of the same Internet meme, in accordance with some implementations.
[0011] Figure 5 A flow chart of an example method for identifying Internet memes is shown in accordance with various implementations.
[0012] Figure 6 A flow chart illustrating an example method for determining whether instances of digital visual content include the same visual moment in time, in accordance with various implementations.
[0013] Figure 7 Shows Figure 5 Continuation of the flowchart, Figure 7 Example methods for determining whether an instance of digital visual content is an instance of a recognized Internet meme are shown in accordance with various implementations.
[0014] Figure 8 Shows Figure 5 Continuation of the flowchart, Figure 8 Example methods for tracking Internet memes are shown in accordance with various implementations. DETAILED DESCRIPTION
[0015] Reference will now be made in detail to various embodiments of the present subject matter, examples of which are shown in the accompanying drawings. Although various embodiments are discussed herein, it should be understood that they are not intended to be limited to these embodiments. On the contrary, the embodiments presented are intended to cover alternatives, modifications and equivalents that may be included in the spirit and scope of the various embodiments defined by the appended claims. In addition, in this specific embodiment, many specific details are set forth in order to provide a thorough understanding of the embodiments of the present subject matter. However, embodiments may be practiced without these specific details. In other cases, well-known methods, processes, components and circuits are not described in detail to avoid unnecessarily obscuring the various aspects of the described embodiments.
[0016] Labeling and nomenclature
[0017] Some parts of the following detailed description are presented in terms of processes, logic blocks, processing and other symbolic representations of operations on data within the circuit. These descriptions and representations are the means used by those skilled in the art of data processing to most effectively convey the essence of their work to other technicians in the field. In this application, processes, logic blocks, processing, etc. are conceived as one or more self-consistent processes or instructions that lead to desired results. A process is a process that requires physical manipulation of physical quantities. Although not necessarily, these quantities are typically in the form of electrical or magnetic signals that can be stored, transferred, combined, compared and otherwise manipulated in an electronic device.
[0018] However, it should be kept in mind that all of these terms and similar terms are associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless otherwise explicitly stated, it will be apparent from the following discussion that throughout the description of the embodiments, discussions using terms such as "monitoring," "determining," "identifying," "tracking," "generating," "comparing," "maintaining," "calculating," and the like refer to actions and processes of electronic devices (such as processors, memories, computing systems, mobile electronic devices, and the like, or combinations thereof). The electronic devices manipulate and convert data represented as physical (electronic and / or magnetic) quantities within registers and memories of the electronic devices into other data similarly represented as physical quantities within memories or registers or other such information storage, transmission, processing, or display components of the electronic devices.
[0019] The embodiments described herein may be discussed in the general context of processor-executable instructions, such as program modules, residing on some form of non-transitory processor-readable media, executed by one or more computers or other devices. Typically, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of the program modules may be combined or distributed as desired.
[0020] In the accompanying drawings, a single block may be described as performing one or more functions; however, in actual practice, the one or more functions performed by the block may be performed in a single component or across multiple components, and / or may be performed using hardware, using software, or using a combination of hardware and software. In order to clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, logic, circuits, and steps have been generally described in terms of their functions. Whether such functions are implemented as hardware or software depends on the specific application and design constraints imposed on the entire system. Technicians can implement the described functions in different ways for each specific application, but such implementation decisions should not be interpreted as resulting in a departure from the scope of the present disclosure. In addition, the example fingerprint sensing system and / or mobile electronic device described herein may include components other than those shown including known components.
[0021] The various techniques described herein may be implemented in hardware, software, firmware, or any combination thereof, unless explicitly described as being implemented in a particular manner. Any features described as modules or components may also be implemented together in an integrated logic device, or separately as discrete but interoperable logic devices. If implemented in software, the technology may be implemented at least in part by a non-transitory processor-readable storage medium comprising instructions that, when executed, perform one or more of the methods described herein. The non-transitory processor-readable data storage medium may form part of a computer program product, which may include packaging materials.
[0022] Non-transitory processor-readable storage media may include random access memory (RAM) such as synchronous dynamic random access memory (SDRAM), read-only memory (ROM), non-volatile random access memory (NVRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, other known storage media, etc. Additionally or alternatively, these techniques may be implemented at least in part through a processor-readable communication medium that carries or communicates code in the form of instructions or data structures and can be accessed, read, and / or executed by a computer or other processor.
[0023] The various embodiments described herein may be performed by one or more processors, such as one or more motion processing units (MPUs), sensor processing units (SPUs), main processors or their cores, digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), application-specific instruction set processors (ASIPs), field programmable gate arrays (FPGAs), programmable logic controllers (PLCs), complex programmable logic devices (CPLDs), discrete gate or transistor logic, discrete hardware components, or any combination thereof, or other equivalent integrated or discrete logic circuit systems, designed to perform the functions described herein. The term "processor" as used herein may refer to any of the aforementioned structures or any other structure suitable for implementing the technology described herein. As used in this specification, the term "processor" may refer to substantially any computing processing unit or device, including but not limited to single-core processors; single processors with software multi-threaded execution capabilities; multi-core processors; multi-core processors with software multi-threaded execution capabilities; multi-core processors using hardware multi-threading technology; parallel platforms; and parallel platforms with distributed shared memory. In addition, the processor may utilize nanoscale architectures such as, but not limited to, transistors, switches, and gates based on molecules and quantum dots in order to optimize space usage or improve the performance of user devices. A processor may also be implemented as a combination of computing processing units.
[0024] In addition, in some aspects, the functionality described herein may be provided in a dedicated software module or hardware module configured as described herein. In addition, these techniques may be fully implemented in one or more circuits or logic elements. A general purpose processor may be a microprocessor, but in an alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices such as a combination of an SPU / MPU and a microprocessor, a plurality of microprocessors, one or more microprocessors in combination with an SPU core, an MPU core, or any other such configuration.
[0025] Overview
[0026] The discussion begins with a description of an example computer system on which embodiments of the invention may be implemented. An example system and example components for identification and tracking of Internet memes are then described. Example operations for identification and tracking of Internet memes are then described.
[0027] The dramatic increase in Internet access has significantly expanded participation in social networks and social media platforms as a means of sharing information. An Internet meme is an example of information shared over the Internet (typically using social networks and social media platforms). An Internet meme is a piece of digital media (e.g., a digital image or digital video clip) that is combined with a text word or phrase to convey or represent a specific theme or meaning. Internet memes can be used to promote or share different types of information including humor, news, entertainment, etc. The goal of many Internet meme creators and promoters is to make an Internet meme go "viral," where the sharing of a viral Internet meme surges rapidly in a short period of time.
[0028] Currently, it is difficult to measure or determine the virality of an Internet meme. In particular, conventional methods for measuring the degree of virality of an Internet meme are typically based on subjective analysis performed after the Internet meme has gained a certain degree of recognition or notoriety. In addition, conventional methods for measuring the virality of an Internet meme are typically platform-specific and cannot reflect the degree of virality of the Internet meme on other platforms. For example, Internet memes are often shared across multiple social networks and social media platforms. As the sharing of Internet memes proliferates across multiple social networks and social media platforms, the ability to track and measure instances of Internet memes collapses due to the proliferating diffuse nature of the proliferating.
[0029] The embodiments described herein provide methods and systems for identifying and tracking Internet memes. As used herein and for the purposes of the embodiments described, Internet memes are defined as fragments (e.g., images or short video clips) of digital visual media content including captions (e.g., text words, phrases, emoticons, or other types of visual emphasis). The identification of Internet memes includes determining the underlying fragments of digital visual media content, which are also referred to herein as "moments", wherein moments can be used as templates for creating different iterations of Internet memes. In other words, Internet memes include underlying moments illustrated using different visual emphasis, wherein each iteration of the moment is an iteration of an Internet meme. An example of an Internet meme is an image of an actor who presents a condescending look in his or her movie role with different text captions, wherein each different instance of the text caption is a different iteration of the same Internet meme.
[0030] Identification of Internet memes allows for objective measurement of the proliferation of Internet memes to be tracked, which can be used to objectively measure the virality of Internet memes. Such information will provide creators and promoters of Internet memes with insight into the characteristics of Internet memes, thereby enabling creators and promoters to understand the effectiveness and reach of particular Internet memes.
[0031] According to various embodiments, methods and systems for identifying and tracking Internet memes are provided. In the method for identifying Internet memes, multiple sources of digital visual content including visual moments and captions are monitored. A determination is made as to whether instances of the digital visual content include the same visual moment. In the event that the instances of the digital visual content include the same visual moment, the instances of the digital visual content including the same visual moment are identified as similar digital visual content. Each similar instance of the digital visual content is tracked. In the event that the total number of instances of the similar digital visual content exceeds an Internet meme threshold, the similar digital visual content is identified as an Internet meme, wherein the same visual moment is a root visual moment and each caption corresponds to a different iteration of the Internet meme.
[0032] Embodiments described herein are about identifying Internet memes by determining whether digital visual content is visually identical or similar. In one embodiment, the digital visual content is a digital image. In other embodiments, the digital visual content is a video file such as a graphics interchange format (GIF) file. It should be understood that other types of pre-recorded media files, such as other types of images and video files, can be used herein. In addition, it should be understood that any type of media file format can be used according to the described embodiment, including but not limited to GIF, WebM, WebP, MPEG-4 (MP4), animated portable network graphics (APNG), motion JPEG, Flash video (FLV), Windows Media video, M4V, etc. It should be understood that the video content item can (for example, by HTML5 video element or Flash video element) loop to automatically repeat. In some embodiments, the video content item is a short-form loop video file (for example, less than 60 seconds).
[0033] In some implementations, the plurality of sources being monitored include known websites and social media platforms such as Facebook, Instagram, Gfycat, Twitter, etc. In some implementations, instances of digital visual content identified as Internet memes are received for comparison with instances of digital visual content from the monitored websites and social media platforms.
[0034] According to some embodiments, perceptual hashing is used to determine the similarity of instances of digital visual content. As used herein, perceptual hashing refers to an algorithm or function applied to a media file to generate a fingerprint or signature for the media file, wherein the greater the similarity between two media files, the greater the similarity between the perceptual hashes used for the two media files. In some embodiments, the output of the perceptual hash function is a string or sequence, which is referred to as a perceptual hash in this article. For example, two images that are visually similar to a person will have similar perceptual hashes, while two images that are visually very different from a person will have less similar perceptual hashes. Perceptual hashing can be contrasted with cryptographic hashing, in which a small difference in the input can completely change the output. Perceptual hashing is useful for determining whether two images are similar. According to the described embodiments, example perceptual hashing functions include phash and dhash, but there are other types of perceptual hashing functions and other types of perceptual hashing functions can be used herein. In various embodiments, perceptual hashing is utilized in the context of a video file by generating a perceptual hash of a frame of the video file.
[0035] According to various embodiments described herein, similarity between media files (e.g., frames of an image or video) is determined by comparing perceptual hashes for the media files. An edit distance (also referred to herein as "distance") between the perceptual hashes is determined. As used herein, edit distance refers to a measure of the difference between two sequences or strings. For example, edit distance is a measure of the degree of difference between two strings or sequences from each other by counting the minimum number of operations required to convert one string into another. According to the described embodiments, various types of distance determinations may be utilized, including but not limited to Levenshtein distance, Hamming distance, Damerau-Levenshtein distance, and longest common sequence (LCS).
[0036] In some embodiments, to determine whether the instances of digital visual content include the same visual moment, a perceptual hash is generated for each instance of digital visual content. The perceptual hashes for each instance of digital visual content are compared to determine a distance between the perceptual hashes. In the event that the distance between the two perceptual hashes satisfies a similarity threshold, it is determined that the instances of digital visual content corresponding to the perceptual hashes include the same visual moment. In some embodiments, in the event that the distance between the two perceptual hashes satisfies the similarity threshold, the instances of digital visual content corresponding to the perceptual hashes are identified as similar digital visual content.
[0037] In some embodiments, once an Internet meme is identified, the instance of the digital visual content may be compared to the Internet meme to determine whether the instance of the digital visual content is also an Internet meme. In the event that the instance of the digital visual content meets a threshold of similarity to the Internet meme, the instance of the digital visual content is identified as an instance of the Internet meme. In some embodiments, it is determined whether the instance of the Internet meme is identical to another instance of the Internet meme. In the event that the instance of the Internet meme is not identical to another instance of the Internet meme, it is determined that the instance of the Internet meme is a new iteration of the Internet meme. In the event that the instance of the Internet meme is identical to another instance of the Internet meme, it is determined that the instance of the Internet meme is another instance of an existing iteration of the Internet meme.
[0038] According to various embodiments, metrics corresponding to each instance of an Internet meme are tracked for use in generating an objective analysis for determining virality of an Internet meme. In some embodiments, instances of each Internet meme are tracked at multiple sources. Each iteration of the Internet meme is tracked. A reach of each iteration of the Internet meme is determined, wherein the reach of each iteration of the Internet meme corresponds to engagement with each iteration of the Internet meme. In some embodiments, a viral score is calculated for each Internet meme, the viral score comprising a first component associated with the reach of each iteration of the Internet meme and a second component associated with the number of iterations of each Internet meme.
[0039] As described above, identification and tracking of Internet memes are important for determining the reach and effectiveness of Internet memes. Providing an objective measurement of the virality of a meme is important for facilitating the creation and promotion of Internet memes. In particular, providing measurable data about the volume and exposure of Internet memes improves the accuracy of the characteristics of Internet memes, providing previously unavailable insights for Internet meme creators and promoters. Therefore, embodiments of the present invention greatly surpass conventional methods of identification and tracking of Internet memes. Furthermore, embodiments of the present invention go far beyond simply using computers to perform identification and tracking of Internet memes. Rather, embodiments of the present invention specifically detail a novel process rooted in computer technology that utilizes objective identification and tracking of Internet memes across multiple websites and social media platforms to determine an objective determination of the virality of an Internet meme.
[0040] Example Computer System and Communication Network
[0041] Turning now to the accompanying drawings, Figure 1A is a block diagram of an example computer system 100 upon which embodiments of the present invention may be implemented. Figure 1A One example of one type of computer system 100 (eg, a computer system) that can be used in accordance with or to implement various embodiments discussed herein is shown.
[0042] It should be understood that Figure 1A The computer system 100 is merely an example, and the embodiments described herein may be operated on or within a number of different computer systems, including but not limited to general-purpose networked computer systems, embedded computer systems, mobile electronic devices, smart phones, server devices, client devices, various intermediate devices / nodes, stand-alone computer systems, media centers, handheld computer systems, multimedia devices, etc. In some embodiments, Figure 1AThe computer system 100 is well suited to having a peripheral tangible computer-readable storage medium 102, such as, for example, an electronic flash data storage device, a floppy disk, a compact disk, a digital versatile disk, other disk-based storage device, a universal serial bus "thumb" drive, a removable memory card, etc., coupled to the computer system 100. Tangible computer-readable storage media are non-transitory in nature.
[0043] Figure 1A The computer system 100 includes an address / data bus 104 for communicating information and a processor 106A coupled to the bus 104 for processing information and instructions. Figure 1A As depicted in , computer system 100 is also well suited for a multi-processor environment where multiple processors 106A, 106B, and 106C are present. Conversely, computer system 100 is also well suited for having a single processor, such as, for example, processor 106A. Processors 106A, 106B, and 106C may be any of a variety of types of microprocessors. Computer system 100 also includes data storage features such as computer usable volatile memory 108 (e.g., random access memory (RAM)) coupled to bus 104 for storing information and instructions for processors 106A, 106B, and 106C. Computer system 100 also includes computer usable non-volatile memory 110 (e.g., read-only memory (ROM)) coupled to bus 104 for storing static information and instructions for processors 106A, 106B, and 106C. Also present in computer system 100 is a data storage unit 112 (e.g., a magnetic disk or optical disk and a magnetic disk drive) coupled to bus 104 for storing information and instructions. Computer system 100 also includes an alphanumeric input device 114 coupled to bus 104 for transmitting information and command selections to processor 106A or processors 106A, 106B, and 106C, the alphanumeric input device 114 including alphanumeric and function keys. Computer system 100 also includes a cursor control device 116 coupled to bus 104 for transmitting user input information and command selections to processor 106A or processors 106A, 106B, and 106C. In one embodiment, computer system 100 also includes a display device 118 coupled to bus 104 for displaying information.
[0044] Still refer to Figure 1A , Figure 1AThe display device 118 may be a liquid crystal device (LCD), a light emitting diode display (LED) device, a cathode ray tube (CRT), a plasma display device, a touch screen device, or other display device suitable for creating graphic images and alphanumeric characters recognizable to a user. The cursor control device 116 allows a computer user to dynamically express the movement of a visible symbol (cursor) on the display screen of the display device 118 and indicates the user's selection of selectable items displayed on the display device 118. Many implementations of the cursor control device 116 are known in the art, including a trackball, a mouse, a touch pad, a touch screen, a joystick, or special keys on the alphanumeric input device 114 that can express movement in a given direction or displacement pattern. Alternatively, it will be understood that the cursor can be directed and / or activated via input from the alphanumeric input device 114 using special keys and key sequence commands. The computer system 100 is also well suited for having the cursor directed by other means such as, for example, voice commands. In various embodiments, the alphanumeric input device 114, the cursor control device 116, and the display device 118, or any combination thereof (e.g., a user interface selection device), may operate together to provide, under the direction of a processor (e.g., processor 106A or processors 106A, 106B, and 106C), a graphical user interface (GUI) 130. The GUI 130 allows a user to interact with the computer system 100 through graphical representations presented on the display device 118 by interacting with the alphanumeric input device 114 and / or the cursor control device 116.
[0045] The computer system 100 also includes an I / O device 120 for coupling the computer system 100 with an external entity. For example, in one embodiment, the I / O device 120 is a modem for implementing wired or wireless communication between the computer system 100 and an external network (such as but not limited to the Internet). In one embodiment, the I / O device 120 includes a transmitter. The computer system 100 can communicate with the network by transmitting data via the I / O device 120.
[0046] Still refer to Figure 1A, depicting various other components for computer system 100. Specifically, operating system 122, applications 124, modules 126, and data 128, when present, are generally shown as residing in one or some combination of computer usable volatile memory 108 (e.g., RAM), computer usable non-volatile memory 110 (e.g., ROM), and data storage unit 112. In some embodiments, all or portions of various embodiments described herein are stored, for example as applications 124 and / or modules 126, in memory locations of computer usable volatile memory 108, computer readable storage media within data storage unit 112, peripheral computer readable storage media 102, and / or other tangible computer readable storage media.
[0047] Figure 1B An example communications network 150 is shown upon which embodiments described herein may be implemented. Figure 1B Computer system 100, social media platform 152, website 154, and internet meme repository 156 are shown, all of which are communicatively coupled via network 150. It should be understood that social media platform 152, website 154, and internet meme repository 156 may be maintained and served by computer system 100 or another instance of a computer system.
[0048] In one embodiment, the social media platform 152, the website 154, and the Internet meme repository 156 are monitored by the computer system 100. The computer system 100 monitors the social media platform 152, the website 154, and the Internet meme repository 156 to obtain digital visual content including visual moments and captions. The visual moment can be any type of file (e.g., an image file or a video file) that can be presented on an electronic device. It should be understood that any type of media file format can be used according to the described embodiments, including but not limited to Graphics Interchange Format (GIF), WebM, WebP, MPEG-4 (MP4), Animated Portable Network Graphics (APNG), Motion JPEG, Flash Video (FLV), Windows Media Video, M4V, etc.
[0049] Example system for identification and tracking of Internet memes
[0050] Figure 2is a block diagram illustrating a system 200 for identifying and tracking Internet memes, according to an embodiment. The system 200 is configured to perform a novel process of identifying changes and tracking Internet memes that must be rooted in computer technology to overcome problems that arise particularly in the field of analysis of digital video content items. In one embodiment, the system 200 provides Internet meme determination 210 for digital visual content 202. The system 200 includes a digital visual content monitor 204, an Internet meme determiner 206, and an analytics tracker 208. The system 200 can be implemented by a computer system (e.g., computer system 100) to access digital visual content from a known source over the Internet to identify and track Internet memes. Although in Figure 2 204, the digital video content item digital visual content monitor 204, the Internet meme determiner 206, and the analytics tracker 208 are shown as separate components, but the digital video content item digital visual content monitor 204, the Internet meme determiner 206, and the analytics tracker 208 may be implemented as a single component or any other combination of components for performing the described functions.
[0051] exist Figure 2 , digital visual content 202 is accessed at a digital visual content monitor 204. In one embodiment, the digital visual content 202 is provided by a user via a client computer system. In other embodiments, the digital visual content 202 is received in response to the digital visual content monitor accessing a known source of digital visual content (such as a website, a social media network, an Internet meme repository) or other source of digital visual content. It should be understood that the digital visual content 202 can be accessed or received using different methods and in response to different conditions.
[0052] In some embodiments, digital visual content 202 is an image file. In some embodiments, digital visual content 202 is a video file such as a GIF file. Although the embodiments described herein are related to GIF files, it should be understood that other types of pre-recorded media files (such as other types of video files) can be used herein. In addition, it should be understood that any type of media file format can be used according to the described embodiments, including but not limited to GIF, WebM, WebP, MPEG-4 (MP4), APNG, motion JPEG, Flash video (FLV), Windows Media video, M4V, etc. It should be understood that digital visual content 202 can (for example, by HTML5 video element or Flash video element) loop to automatically repeat. In some embodiments, digital visual content 202 is a short-form loop video file (for example, less than 60 seconds).
[0053] Reference Figure 2, the internet meme determiner 206 processes the identification of whether the digital visual content 202 is an internet meme. In some embodiments, the internet meme determiner 206 determines whether the digital visual content 202 includes the same visual moment as other instances of digital visual content. In the event that the instances of digital visual content include the same visual moment, it is determined that the digital visual content 202 is identified as digital visual content similar to the corresponding digital visual content.
[0054] The Internet meme determiner 206 is also configured to track each instance of similar digital visual content. This tracking enables a determination of whether an instance of a new digital visual content (or an instance of digital visual content that has not been previously identified as an Internet meme) is an Internet meme. In the event that the total number of instances of similar digital visual content exceeds an Internet meme threshold (e.g., 10 instances), the similar digital visual content is identified as an Internet meme, wherein the same visual moment is a root visual moment and each caption corresponds to a different iteration of the Internet meme. The Internet meme determiner 206 is configured to generate an Internet meme determination 210 indicating whether the digital visual content 202 is an Internet meme.
[0055] According to some embodiments, the Internet meme determiner 206 is configured to identify new Internet memes in real time as they are created. Internet memes for this purpose are images or videos (visual moments) that become templates for Internet users to iterate over and over again using different captions. Root visual moments (e.g., images or videos) are identified, where the identification is based on visual similarity. In some embodiments, a machine learning engine that generates visual tensors is used. Each tensor has a distance metric so that the distance between any two images can be calculated. The visual tensor has the following properties:
[0056] ■The same binary files will have the same tensors;
[0057] ■ Tensors for visually identical content saved in different formats will have identical or nearly identical tensors; and
[0058] ■ Small visual changes - such as changing resolution, changing color, or changing part of the content - will result in very low distance scores between each other and can be confidently called the same tensor.
[0059] A determination is then made as to whether the same root visual moment is repeatedly iterated with changes. In one embodiment, this determination detects whether the visual content is the same but the text is different. In another embodiment, this determination detects whether the visual content is nearly the same but moderately different. Mathematically, this means that clusters of tensors have a distance measure that is less than X but greater than Y, where X and Y are chosen such that new changes are likely to be greater than X, but unrelated content is likely to be greater than Y.
[0060] In other implementations, the Internet meme determiner 206 is configured to identify an Internet meme by comparing the digital visual content 202 to determine whether the digital visual content 202 is an instance of a previously identified Internet meme.
[0061] Figure 3 is a block diagram illustrating an Internet meme determiner 206 of a system 200 for identifying Internet memes according to an embodiment. In one embodiment, the Internet meme determiner 206 receives a digital visual content 202. A perceptual hash generator 302 generates a perceptual hash for the digital visual content 202. The perceptual hash generator 302 is configured to generate a perceptual hash for the digital visual content 202. In some embodiments, when the digital visual content 202 is a video file, a plurality of perceptual hashes corresponding to frames of the video file may be generated. Perceptual hashing as used herein refers to an algorithm or function applied to a media file to generate a fingerprint or signature of the media file, wherein the greater the similarity between two media files, the greater the similarity between the perceptual hashes of the two media files. In some embodiments, the output of the perceptual hash function is a string or sequence, which is referred to herein as a perceptual hash. For example, two images that are visually similar to a person will have similar perceptual hashes, while two images that are visually very different from a person will have less similar perceptual hashes. Perceptual hashing can be contrasted to cryptographic hashing, where small differences in the input can drastically change the output. Perceptual hashing is useful for determining whether two images are similar. According to the described embodiments, example perceptual hash functions include phash and dhash, but other types of perceptual hash functions exist and may be used herein. In various embodiments, perceptual hashing is used in the context of a video file by generating perceptual hashes for frames of the video file.
[0062] According to various embodiments described herein, similarities between media files (e.g., frames of images or videos) are determined by comparing perceptual hashes for media files. Perceptual hash comparator 304 is configured to compare perceptual hashes of digital visual content 202 with previously generated perceptual hashes corresponding to other digital visual content (some of which can be identified as Internet memes) stored in perceptual hash library 306. Edit distances (also referred to herein as "distances") between perceptual hashes are determined. As used herein, edit distances refer to the measurement of differences between two sequences or strings. For example, edit distances measure the degree of difference between two strings or sequences by counting the minimum number of operations required to convert a string into another string. Various types of distances can be used to determine according to the described embodiments, including but not limited to Levenshtein distance, Hamming distance, Damerau-Levenshtein distance, and longest common sequence (LCS).
[0063] An edit distance between the perceptual hash for the digital visual content 202 and the perceptual hash of the perceptual hash library 306 is determined. In the event that the edit distance between the two perceptual hashes satisfies a similarity threshold, the instances of digital visual content corresponding to the perceptual hashes are determined to include the same visual moment. It should be appreciated that the similarity threshold allows for smaller differences in the digital visual content and can be adjusted according to various confidence levels. For example, if the two perceptual hashes are the same, it is determined that the instances of digital visual content are the same and refer to the same image. If the two perceptual hashes are not the same but meet the similarity threshold, the instances of digital visual content may include the same underlying visual moment but include different captions and are determined to be different instances of the same digital visual content.
[0064] As described above, if it is determined that the number of different instances of the same digital visual content exceeds an internet meme threshold (e.g., 10 instances), the similar digital visual content is identified as an internet meme, where the same visual moment is a root visual moment and each caption corresponds to a different iteration of the internet meme.
[0065] Reference Figure 4A and Figure 4B , shows examples of different iterations of the same Internet meme according to some embodiments. Figure 4A As shown in FIG. 4 , a digital image 400 is shown, wherein an underlying moment 402 of the digital image 400 includes an actor playing a character in a movie and a caption 404 stating the phrase “please tell me more”. Figure 4B, a digital image 410 is shown, wherein an underlying moment 412 of the digital image 410 includes an actor playing a character and a caption 414 narrating the phrase "dogo on". The underlying moments 402 and 412 shown are identical and are the same underlying moment, while the captions 404 and 414 are different. Thus, according to the described embodiment, the digital images 400 and 410 are different instances of the same Internet meme.
[0066] Refer again Figure 2 , analytics tracker 208 is configured to track every instance of every Internet meme at a known source, including tracking of every iteration of every Internet meme. Analytics 212 are generated by analytics tracker 208, wherein analytics 212 includes objective data quantifying the reach and virality of each Internet meme (and each iteration of each Internet meme).
[0067] In some embodiments, after an Internet meme has been identified, each individual Internet meme is then tracked and recorded over time. These measurements include the total number of iterations of the Internet meme, the reach of each iteration of the Internet meme, and an overall score for each Internet meme based on the number of iterations and reach of the Internet meme.
[0068] Total number of iterations of an Internet meme: Record when each iteration of an Internet meme occurs. In some embodiments, an iteration is counted only when it is generated by a different social media account than the previous iteration. Iterations of Internet memes can include Internet memes that are text-generated as images or videos. For example, someone uses an Internet meme template with an image or video and changes it to the Internet meme template and makes a derivative. An example can be using an image of the President and the Speaker of the House in the State of the Union address (one image (roughly)) and people modify the image with different text captions or effects. Iterations of Internet memes can also include related Internet memes that are copied in style but are not from the exact same source image or video of the original meme. Examples of this can be different screenshots of exaggerated hand movements of politicians when giving speeches or extended screaming scenes of actors in mini-series.
[0069] Range for each iteration. In some embodiments, the range is calculated using the following steps: 1) engagement measures from each social media platform, e.g., likes on Facebook, Twitter, and Instagram, likes on Reddit, number of comment replies on 4chan; 2) total number of actual views on an Internet meme repository (e.g., Gfycat / Memento or others), when data is available; and 3) normalizing each measure using a weighted ranking. For example, for a set of existing Gfycat posts on twitter, read the number of retweets and likes and correlate it with the average retweets per view and the average likes per view.
[0070] An overall score is calculated combining iterations and ranges. The purpose of the score is to objectively measure cultural influence, so it is unlikely for a high-impact meme to have a low score, and it is unlikely for a low-impact meme to have a high score. The formula can be stated in the most general way to satisfy these conditions:
[0071] ●For one iteration: social_score = f(Likes, Shares, Views), where f is some strictly increasing monotonic function.
[0072] ●For the set of all iterations of a meme: total_score = f({a,b,c,...}), where {a,b,c,...} is the N-sized set of social_scores for each iteration of the meme and f is some monotonic function that (a) strictly increases with the input (Likes, Shares, Views) of each social_score and (b) strictly increases with N (the size of the set), which by definition is the total number of iterations of the meme.
[0073] • An example formula that satisfies these conditions is a simple sum of the scores for each iteration, with each score being a simple sum of the number of social engagements.
[0074] As presented above, identification and tracking of Internet memes are important for determining the reach and effectiveness of Internet memes. Providing an objective measurement of the virality of a meme is important for facilitating the creation and promotion of Internet memes. In particular, providing measurable data about the volume and exposure of Internet memes improves the accuracy of the characteristics of Internet memes, thereby providing previously unavailable insights to Internet meme creators and promoters. Therefore, embodiments of the present invention greatly surpass conventional methods of identification and tracking of Internet memes. In addition, embodiments of the present invention go far beyond simply using computers to perform identification and tracking of Internet memes. Instead, embodiments of the present invention specifically detail a novel process rooted in computer technology that utilizes objective identification and tracking of Internet memes across multiple websites and social media platforms to determine an objective determination of the virality of an Internet meme.
[0075] Example operation of identifying and tracking Internet memes
[0076] The following discussion sets forth in detail the operation of some example methods of operation of the embodiments. Figures 5 to 8, flowcharts 500, 600, 700, and 800 illustrate example processes used by various embodiments. Flowcharts 500, 600, 700, and 800 include some processes performed by a processor under the control of computer-readable instructions and computer-executable instructions in various embodiments. In this manner, in various embodiments, the processes described herein and described in conjunction with the flowcharts are implemented using a computer or can be implemented using a computer. Computer-readable instructions and computer-executable instructions can reside in any tangible computer-readable storage medium. Some non-limiting examples of tangible computer-readable storage media include random access memory, read-only memory, magnetic disks, solid-state drives / "disks," and optical disks, any or all of which can be used with a computer environment (e.g., computer system 100). Computer-readable instructions and computer-executable instructions residing on tangible computer-readable storage media are used to control or operate, for example, in conjunction with one or some combination of processors of a computer environment and / or a virtualized environment. It should be understood that the processor can be physical or virtual or some combination (it should also be understood that the virtual processor is implemented on physical hardware). Although specific processes are disclosed in the flowcharts, such processes are examples. That is, embodiments are well suited to performing various other processes or variations of the processes enumerated in the flowcharts. Likewise, in some embodiments, the processes in flowcharts 500, 600, 700, and 800 may be performed in an order different from the order presented and / or not all of the processes described in flowcharts 500, 600, 700, and 800 may be performed. It should also be understood that the processes described in flowcharts 500, 600, 700, and 800 may be implemented in hardware or a combination of hardware and firmware and / or software provided by computer system 100.
[0077] Figure 5 A flow chart 500 of an example method for identifying an Internet meme is shown according to various embodiments. At process 510, multiple sources of digital visual content including visual moments and captions are monitored. According to various embodiments, the multiple sources include websites and social media sites. In one embodiment, as shown at step 515, example digital visual content is received. In one embodiment, the digital visual content includes images. In another embodiment, the digital visual content includes a video. In one embodiment, the video is a short-form looping video content item.
[0078] At process 520, it is determined whether the instances of digital visual content include the same visual moment. If it is determined that the instance digital visual content does not include the same visual moment, flowchart 500 ends as shown at process 525. Alternatively, if it is determined that the instance digital visual content does include the same visual moment, flowchart 500 proceeds to process 530. In one embodiment, according to Figure 6Flowchart 600 of execution process 520 .
[0079] Figure 6 A flowchart 600 of an example method for determining whether instances of digital visual content include the same visual moment in accordance with various embodiments is shown. At process 610, a perceptual hash is generated for each instance of digital visual content. At process 620, the perceptual hash for each instance of digital visual content is compared to determine a distance between the perceptual hashes. At process 630, it is determined whether the distance satisfies a similarity threshold. In the event that the distance between the two perceptual hashes does not satisfy the similarity threshold, flowchart 600 ends as shown at process 635. In the event that the distance between the two perceptual hashes satisfies the similarity threshold, as shown at process 640, it is determined that the instances of digital visual content corresponding to the perceptual hashes include the same visual moment.
[0080] Reference Figure 5 At process 530, when the instances of digital visual content include the same visual moment, the instances of digital visual content including the same visual moment are identified as similar digital visual content. In one embodiment, when the distance between the two perceptual hashes satisfies a similarity threshold, as shown at process 630, the instances of digital visual content corresponding to the perceptual hashes are identified as similar digital visual content.
[0081] At process 540, each instance of similar digital visual content is tracked. In one embodiment, as shown at process 545, a count of each instance of similar digital visual content is maintained. At process 550, it is determined whether the number of instances of similar digital visual content exceeds an Internet meme threshold. In the event that the total number of instances of similar digital visual content does not exceed the Internet meme threshold, flowchart 500 returns to process 540. In the event that the total number of instances of similar digital visual content exceeds the Internet meme threshold, as shown at process 560, the similar digital visual content is identified as an Internet meme, wherein the same visual moment is a root visual moment and each caption corresponds to a different iteration of the Internet meme.
[0082] Figure 7A continuation of flowchart 500 of an example method for determining whether an instance of digital visual content is an instance of an identified Internet meme is shown in accordance with various embodiments. At process 562, the instance of digital visual content is compared to an Internet meme. At process 564, it is determined whether the instance of digital visual content meets a similarity threshold for the Internet meme. If it is determined that the instance of digital visual content does not meet the similarity threshold, flowchart 500 ends as shown at process 566. If it is determined that the instance of digital visual content does meet the similarity threshold, as shown at process 568, the instance of digital visual content is identified as an instance of an Internet meme. In one embodiment, it is determined whether the instance of the Internet meme is the same as another instance of the Internet meme. In the case where the instance of the Internet meme is not the same as another instance of the Internet meme, it is determined that the instance of the Internet meme is a new iteration of the Internet meme. In the case where the instance of the Internet meme is the same as another instance of the Internet meme, it is determined that the instance of the Internet meme is another instance of an existing iteration of the Internet meme.
[0083] Figure 8 A continuation of flowchart 500 of an example method for tracking Internet memes is shown in accordance with various embodiments. At process 572, each instance of an Internet meme is tracked at a plurality of sources. At process 574, each iteration of the Internet meme is tracked. At process 576, a scope of each iteration of the Internet meme is determined, wherein the scope of each iteration of the Internet meme corresponds to engagement with each iteration of the Internet meme. In one embodiment, as shown at process 578, a viral score is calculated for each Internet meme, the viral score comprising a first component associated with the scope of each iteration of the Internet meme and a second component associated with the number of iterations of each Internet meme.
[0084] in conclusion
[0085] The examples set forth herein are presented to best explain, describe specific applications and thereby enable those skilled in the art to make and use the embodiments of the described examples. However, those skilled in the art will recognize that the foregoing description and examples are presented for purposes of illustration and example only. Many aspects of the different example embodiments described above may be combined into new embodiments. The set forth description is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Rather, the specific features and actions described above are disclosed as example forms for implementing the claims.
[0086] References throughout this document to "one embodiment," "certain embodiments," "embodiments," "various embodiments," "some embodiments," or similar terms mean that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least one embodiment. Thus, the appearances of such phrases in various places throughout the specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics of any embodiment may be combined with one or more other features, structures, or characteristics of one or more other embodiments in any suitable manner without limitation.
[0087] The solution of the present invention also includes:
[0088] (1) A method for identifying Internet memes, the method comprising:
[0089] monitoring multiple sources of digital visual content including visual moments and captions;
[0090] determining whether the digital visual content instances include the same visual moment;
[0091] In case the digital visual content instances include the same visual moment, identifying the digital visual content instances including the same visual moment as similar digital visual content;
[0092] tracking each instance of the similar digital visual content; and
[0093] In case the total number of instances of the similar digital visual content exceeds an internet meme threshold, the similar digital visual content is identified as an internet meme, wherein the same visual moment is a root visual moment and each caption corresponds to a different iteration of the internet meme.
[0094] (2) The method of (1), wherein determining whether the instances of digital visual content include the same visual moment comprises:
[0095] generating a perceptual hash for each instance of digital visual content;
[0096] comparing the perceptual hashes of each instance of the digital visual content to determine a distance between the perceptual hashes; and
[0097] In case the distance between two perceptual hashes satisfies a similarity threshold, it is determined that the instance of the digital visual content corresponding to the perceptual hash includes the same visual moment.
[0098] (3) The method according to (2), wherein identifying the instances of digital visual content including the same visual moment as similar digital visual content comprises:
[0099] In case that the distance between the two perceptual hashes satisfies the similarity threshold, the instances of digital visual content corresponding to the two perceptual hashes are identified as similar digital visual content.
[0100] (4) The method of (1), wherein the plurality of sources include websites and social media sites.
[0101] (5) The method according to (1), wherein the digital visual content includes an image.
[0102] (6) The method according to (1), wherein the digital visual content comprises a video.
[0103] (7) The method of (6), wherein the video is a short-form looping video content item.
[0104] (8) The method according to (1), wherein tracking each instance of the similar digital visual content comprises:
[0105] A count of each instance of the similar digital visual content is maintained.
[0106] (9) The method according to (1), further comprising:
[0107] Compare instances of digital visual content to Internet memes; and
[0108] In case the instance of the digital visual content satisfies a similarity threshold, the instance of the digital visual content is identified as an instance of the Internet meme.
[0109] (10) The method according to (9), further comprising:
[0110] determining whether the instance of the Internet meme is the same as another instance of the Internet meme;
[0111] In a case where the instance of the Internet meme is not identical to another instance of the Internet meme, determining that the instance of the Internet meme is a new iteration of the Internet meme; and
[0112] In the event that the instance of the Internet meme is identical to another instance of the Internet meme, it is determined that the instance of the Internet meme is another instance of an existing iteration of the Internet meme.
[0113] (11) The method according to (1), further comprising:
[0114] tracking each instance of the internet meme at the plurality of sources;
[0115] Tracking every iteration of said internet meme; and
[0116] A reach of each iteration of the Internet meme is determined, wherein the reach of each iteration of the Internet meme corresponds to engagement with each iteration of the Internet meme.
[0117] (12) The method according to (11), further comprising:
[0118] A viral score is calculated for each Internet meme, the viral score comprising a first component associated with the scope of each iteration of the Internet meme and a second component associated with the number of iterations of each Internet meme.
[0119] (13) The method according to (1), further comprising:
[0120] Receive the instance digital visual content.
[0121] (14) A non-transitory computer-readable storage medium having a computer-readable program code stored thereon, the computer-readable program code being used to cause a computer system to execute a method for identifying Internet memes, the method comprising:
[0122] monitoring multiple sources of digital visual content including visual moments and captions;
[0123] determining whether instances of digital visual content include the same visual moment;
[0124] In case the instances of the digital visual content include the same visual moment, identifying the instances of the digital visual content including the same visual moment as similar digital visual content;
[0125] tracking each instance of the similar digital visual content; and
[0126] In a case where the total number of instances of the similar digital visual content exceeds an internet meme threshold, the similar digital visual content is identified as an internet meme, wherein the same visual moment is a root visual moment and each caption corresponds to a different iteration of the internet meme.
[0127] (15) The non-transitory computer-readable storage medium of (14), wherein determining whether the instances of the digital visual content include the same visual moment comprises:
[0128] generating a perceptual hash for each instance of digital visual content;
[0129] comparing the perceptual hashes of each instance of the digital visual content to determine a distance between the perceptual hashes; and
[0130] In case the distance between two perceptual hashes satisfies a similarity threshold, it is determined that the instance of the digital visual content corresponding to the perceptual hash includes the same visual moment.
[0131] (16) The non-transitory computer-readable storage medium of (15), wherein identifying the instances of digital visual content including the same visual moment as similar digital visual content comprises:
[0132] In case that the distance between the two perceptual hashes satisfies the similarity threshold, the instances of digital visual content corresponding to the two perceptual hashes are identified as similar digital visual content.
[0133] (17) According to the non-transitory computer-readable storage medium of (14), the method further comprises:
[0134] Compare instances of digital visual content to Internet memes; and
[0135] In case the instance of the digital visual content satisfies a similarity threshold, the instance of the digital visual content is identified as an instance of the Internet meme.
[0136] (18) According to the non-transitory computer-readable storage medium of (17), the method further comprises:
[0137] determining whether the instance of the Internet meme is the same as another instance of the Internet meme;
[0138] In a case where the instance of the Internet meme is not identical to another instance of the Internet meme, determining that the instance of the Internet meme is a new iteration of the Internet meme; and
[0139] In the event that the instance of the Internet meme is identical to another instance of the Internet meme, it is determined that the instance of the Internet meme is another instance of an existing iteration of the Internet meme.
[0140] (19) According to the non-transitory computer-readable storage medium of (14), the method further comprises:
[0141] tracking each instance of the internet meme at the plurality of sources;
[0142] Tracking every iteration of said internet meme;
[0143] determining a reach of each iteration of the Internet meme, wherein the reach of each iteration of the Internet meme corresponds to engagement with each iteration of the Internet meme; and
[0144] A viral score is calculated for each Internet meme, the viral score comprising a first component associated with the scope of each iteration of the Internet meme and a second component associated with the number of iterations of each Internet meme.
[0145] (20) A computer system comprising:
[0146] data storage unit; and
[0147] A processor coupled to the data storage unit, the processor being configured to:
[0148] monitoring multiple sources of digital visual content including visual moments and captions;
[0149] generating a perceptual hash for each instance of digital visual content;
[0150] comparing the perceptual hashes for each instance of the digital visual content to determine a distance between the perceptual hashes;
[0151] In the case where the distance between the two perceptual hashes satisfies a similarity threshold, determining that the instances of digital visual content corresponding to the perceptual hashes include a same visual moment, and identifying the instances of digital visual content corresponding to the perceptual hashes as similar digital visual content;
[0152] tracking each instance of the similar digital visual content; and
[0153] In a case where the total number of instances of the similar digital visual content exceeds an internet meme threshold, the similar digital visual content is identified as an internet meme, wherein the same visual moment is a root visual moment and each caption corresponds to a different iteration of the internet meme.
Claims
1. A method for tracking digital memes, the method comprising: monitoring multiple sources of digital visual content including visual moments and captions; compare instances of digital visual content to digital memes; In a case where the instance of the digital visual content satisfies a similarity threshold, identifying the instance of the digital visual content as an instance of the digital meme; determining whether the instance of the digital meme is the same as the existing iteration of the digital meme based on a comparison between the caption of the existing iteration of the digital meme and the caption of the instance of the digital meme; Upon determining that the instance of the digital meme is not identical to the existing iteration of the digital meme based on the instance of the digital meme including captions that are different from captions of the existing iteration of the digital meme, tracking the instance of the digital meme as a new iteration of the digital meme; as well as Upon determining that the instance of the digital meme is the same as the existing iteration of the digital meme based on the instance of the digital meme including captions that are the same as the captions of the existing iteration of the digital meme, tracking the instance of the digital meme as another instance of the existing iteration of the digital meme.
2. The method according to claim 1, further comprising: generating a perceptual hash for an instance of the digital visual content; comparing a perceptual hash for the digital visual content to a perceptual hash for the digital meme; as well as In case the distance between the two perceptual hashes satisfies a similarity threshold, the instance of the digital visual content is identified as an instance of the digital meme.
3. The method according to claim 1, wherein: The digital visual content includes one or more frames of a video.
4. The method according to claim 1, wherein: An instance of the digital meme is tracked as another instance of the existing iteration of the digital meme only if it is generated by a different social media account than the existing iteration of the digital meme.
5. The method according to claim 1, further comprising: tracking each instance of the digital meme at the plurality of sources; Tracking each iteration of said digital meme; as well as A scope of each iteration of the digital meme is determined, wherein the scope of each iteration of the digital meme corresponds to engagement with each iteration of the digital meme.
6. The method according to claim 5, further comprising: A viral score is calculated for each digital meme, the viral score comprising a first component associated with a scope of each iteration of the digital meme and a second component associated with a number of iterations of each digital meme.
7. The method according to claim 5, further comprising: The reach of existing iterations of a digital meme is normalized based on an average of engagement measurements of the digital meme at a particular source of digital visual content.
8. A non-transitory computer readable storage medium having computer readable program code stored thereon, the computer readable program code being configured to cause a computer system to perform a method for identifying digital memes, the method comprising: monitoring multiple sources of digital visual content including visual moments and captions; compare instances of digital visual content to digital memes; In a case where the instance of the digital visual content satisfies a similarity threshold, identifying the instance of the digital visual content as an instance of the digital meme; determining whether the instance of the digital meme is the same as the existing iteration of the digital meme based on a comparison between the caption of the existing iteration of the digital meme and the caption of the instance of the digital meme; Upon determining that the instance of the digital meme is not identical to the existing iteration of the digital meme based on the instance of the digital meme including captions that are different from captions of the existing iteration of the digital meme, tracking the instance of the digital meme as a new iteration of the digital meme; as well as Upon determining that the instance of the digital meme is the same as the existing iteration of the digital meme based on the instance of the digital meme including captions that are the same as the captions of the existing iteration of the digital meme, tracking the instance of the digital meme as another instance of the existing iteration of the digital meme.
9. The non-transitory computer-readable storage medium of claim 8, wherein: The method further comprises: generating a perceptual hash for an instance of the digital visual content; comparing a perceptual hash for the digital visual content to a perceptual hash for the digital meme; and In case the distance between the two perceptual hashes satisfies a similarity threshold, the instance of the digital visual content is identified as an instance of the digital meme.
10. A computer system for tracking digital memes, comprising: a data storage unit including processor-readable instructions; as well as At least one processor coupled to the data storage unit, the at least one processor being configured by the processor-readable instructions to perform operations comprising: monitoring multiple sources of digital visual content including visual moments and captions; compare instances of digital visual content to digital memes; In a case where the instance of the digital visual content satisfies a similarity threshold, identifying the instance of the digital visual content as an instance of the digital meme; determining whether the instance of the digital meme is the same as the existing iteration of the digital meme based on a comparison between the caption of the existing iteration of the digital meme and the caption of the instance of the digital meme; Upon determining that the instance of the digital meme is not identical to the existing iteration of the digital meme based on the instance of the digital meme including captions that are different from the captions of the existing iteration of the digital meme, tracking the instance of the digital meme as a new iteration of the digital meme; and Upon determining that the instance of the digital meme is the same as the existing iteration of the digital meme based on the instance of the digital meme including captions that are the same as the captions of the existing iteration of the digital meme, tracking the instance of the digital meme as another instance of the existing iteration of the digital meme.
