KR102308893B1 - Method for operating contents using recognition meta and service device supporting the same - Google Patents

Abstract

A service apparatus according to an embodiment of the present invention may include a memory and a processor functionally connected to the memory. The processor obtains the content stored in the memory, extracts recognition metas from the content by using at least one recognition technology, and generates a recognition meta map in the form of a map using the extracted recognition metas and stores it in the memory can be set to

Description

Method for operating contents using recognition meta and service device supporting the same}

The present invention relates to content recommendation, and in particular, classifying and storing contents based on the recognition meta, and using the recognition meta to search for content similar to the user's preferred content and content management using the recognition meta that supports the provision of a content recommendation list It relates to a method and a service apparatus supporting the same.

In order to recommend a movie or video in a conventional content management system, methods such as a recommendation based on the characteristics of a video and a recommendation based on a user's viewing history have been used. Among these, the recommendation technique based on the characteristics of the video is performed by setting the characteristics of the video genre, actors, director, synopsis, etc. as parameters, analyzing the similarity based on this, and recommending contents according to the analysis result. do. This conventional content recommendation method simply defines a specific movie as a single word, such as “thriller,” or roughly defines the video characteristics only with basic metadata such as actors and directors. There was a difficult problem. For example, if some of the contents of a movie are "thriller", later some contents are "romance", and the last contents are composed of "comedy", it is not easy to define a genre with some characteristic expressions for the movie. , when a content recommendation operation is performed due to this, as information describing only some characteristics of the content is provided, there is a problem in that the reliability of the recommended information is lowered.

Therefore, in order to solve the above difficulties, in addition to meta such as "thriller", "romance", and "comedy", which are the genre keywords of a movie, recognition-based meta work that can read and store the flow in the video may be required. That is, the cast (person), sound (sound), object (thing), place, situation extracted from the video itself through various recognition technologies, rather than text metadata that records the contents of the synopsis, plot, genre, and cast list. Various recognition-based metadata such as

Korean Patent Laid-Open Patent No. 10-2009-033750, registered on April 06, 2009 (Title: Contents Play List Recommendation Method and Apparatus)

The present invention provides a method for managing content using recognition meta that can effectively store and manage meta data extracted by content content recognition technology, and accurately recommend an image using the recognition meta data, and the method to meet the above needs. It is to provide a service device that supports it.

A service apparatus according to an embodiment of the present invention may include a memory and a processor functionally connected to the memory. The processor obtains the content stored in the memory, extracts recognition metas from the content by using at least one recognition technology, and generates a recognition meta map in the form of a map using the extracted recognition metas and stores it in the memory can be set to

Here, the processor may be configured to generate recognition meta maps for the contents stored in the memory, classify the contents based on the recognition meta maps, and store classification information according to the classification in the memory.

The service device may further include a communication circuit forming a communication channel with the terminal device. In this case, when receiving the content for search from the terminal device, the processor may generate a recognition meta map for search by applying a recognition technology and recognition unit designated for the content for search.

Here, the processor extracts a distribution value of the recognized species information and the recognized species information required for a recognition meta map search from among the information included in the search request information, and the recognized species information is within a certain similarity with the specified distribution value. It may be configured to detect a recognition meta map having a distribution value.

In addition, the service device further includes a communication circuit forming a communication channel with the terminal device, in this case, the processor receives the search request information from the terminal device, and at least one recognition meta map corresponding to the search request information may be configured to detect in the memory, detect at least one content corresponding to the recognition meta map, generate a recommendation list based on the at least one content, and provide it to the terminal device through the communication meeting. .

Meanwhile, the processor may set the number of recognition species to be applied to the recognition meta map differently according to the length of the reproduction time of the content.

The recognition meta map may include information in two-dimensionally arranging recognition metas recognized based on the at least one recognition technology with respect to at least a portion of the reproduction frames of the content.

The processor collects recognition technology and recognition unit information included in the search request information, collects a plurality of recognition meta maps corresponding to the recognition technology and recognition unit information, and collects the plurality of recognition meta maps and the search request information It may be configured to simultaneously compare recognition meta maps for search included in .

A service device operating method according to an embodiment of the present invention includes, by the service device, acquiring contents stored in a memory, extracting recognition metas from the contents using at least one recognition technology, and the extracted recognition meta generating recognition metamaps in the form of a map for each content using have.

The computer recording medium according to an embodiment of the present invention provides an operation for a service device to acquire content stored in a memory, an operation for extracting recognition metas from the content using at least one recognition technology, and an operation for extracting recognition metas from the content using the extracted recognition metas. at least one set to perform an operation of generating a recognition meta map in the form of a map for each content, an operation of classifying contents based on a plurality of recognition meta maps, and an operation of storing classification information according to the classification in the memory; You can store commands.

According to the present invention, by constructing a recognition meta from the contents of the image content to construct a two-dimensional recognition meta map, the characteristic classification of the image content can be processed using relatively little data, and the image analysis and comparison can be processed more quickly support you to do it

In addition, the present invention supports efficient storage space management by compressing and storing the recognition meta map related to image content.

In addition, other effects of the present invention will be mentioned in the following description.

1 is a diagram illustrating an example of a network environment including a content management system according to an embodiment of the present invention.
2 is a diagram showing a schematic form of a service apparatus according to an embodiment of the present invention.
3 is a diagram illustrating an example of a processor of a service device according to an embodiment of the present invention.
4 illustrates an example of a recognition meta map in the form of a two-dimensional dynamic array map according to an embodiment of the present invention.
5 illustrates another example of a recognition meta map in the form of a two-dimensional dynamic array map according to an embodiment of the present invention.
6 is a diagram illustrating an example of a structure for species management of a recognition meta according to an embodiment of the present invention.
7A is a diagram illustrating that it is possible to simultaneously inquire recognition meta maps of four types of recognition meta in a certain axis direction.
7B is a diagram illustrating examples of recognition meta maps for various movie contents.
8 is a diagram illustrating an example of storing a recognition meta map according to an embodiment of the present invention.
9 is a diagram illustrating an example of a method for operating a service device according to an embodiment of the present invention.

Hereinafter, various embodiments of the present invention will be described with reference to the accompanying drawings. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that various modifications, equivalents, and/or alternatives of the embodiments of the present invention are included. In connection with the description of the drawings, like reference numerals may be used for like components.

1 is a diagram illustrating an example of a network environment including a content management system according to an embodiment of the present invention.

Referring to FIG. 1 , a network environment 10 including a content management system according to an embodiment of the present invention may include a network 50 , a terminal device 100 , and a service device 200 .

In the above-described network environment 10 of the present invention, when the terminal device 100 accesses the service device 200 through the network 50 , the service device 200 provides a content use screen designated to the terminal device 100 . In response to the search request information requested by the terminal device 100 , the service device 200 may search for content corresponding to the search request information and provide it to the terminal device 100 . In this process, the service device 200 extracts recognition metas by recognizing the content of each content as a predetermined recognition unit with respect to a plurality of pre-stored content, and uses the extracted recognition metas as a recognition meta map corresponding to the content. can be configured as The service device 200 configures a recognition meta map for search for the content requested by the terminal device 100, searches for data corresponding to the recognition meta map for search in the recognition meta map DB stored in the memory, and results in a search may be provided to the terminal device 100 . The present invention as described above effectively manages source data extracted as various recognition technologies corresponding to image contents in a media meta management system used by a media platform operator to provide content, and utilizes it to configure a content management system.

The network 50 is not only an IP-based wired communication network such as an Internet network, but also various types of wireless networks such as a long term evolution (LTE) network, a mobile communication network such as a WCDMA network, a Wi-Fi network, and combinations thereof. can be made with That is, the network environment 10 related to the content management function (eg, content storage using recognition meta and content recommendation function using stored recognition meta) according to the present invention can be applied to all wired and wireless communication networks without distinction. Specifically, the network 50 may establish a communication channel between the service device 200 and the terminal device 100 . For example, the network 50 may support at least one of 3G, 4G, and 5G wireless mobile communication methods that the service device 200 or the terminal device 100 can operate. Alternatively, the network 50 may establish a communication channel between the terminal device 100 and the service device 200 based on a wire. Such a network 50 should be interpreted as a concept including various wired networks, wireless networks, and combinations thereof that are currently developed and commercialized or that will be developed and commercialized in the future.

The terminal device 100 may access the service device 200 through the network 50 . The terminal device 100 may receive and output a content use screen provided by the service device 200 , and provide search request information to the service device 200 in response to a user input. In this operation, the terminal device 100 provides search request information including information such as specific content or a recognition meta map generated based on the specific content to the service device 200 , and receives the corresponding information from the service device 200 . Recommendation information corresponding to the search request information may be received and output. In this regard, the terminal device 100 includes a communication circuit, a memory, a processor, an input unit, a display, and the like, and when receiving a search request for content stored in the memory from the input unit, it generates a recognition meta map corresponding to the stored content. and may provide the recognition meta map as search request information to the service device 200 . Here, the terminal device 100 provides general metadata of the content, for example, the title of the content, genre information, director or actor information, synopsis information, and the like, to the service device 200 as search request information together with the recognition meta map, and , a recommendation list corresponding to the corresponding information may be received.

The service device 200 may have a communication standby state so that the terminal device 100 can access it through the network 50 . When the terminal device 100 connects, the service device 200 displays a content use screen (or virtual page) for using the content recommendation service provided by the service device 200 to the terminal device 100 ( 100) can be provided. The service device 200 may receive search request information from the terminal device 100 , search for content corresponding to the search request information, configure the searched content as a recommendation list, and provide it to the terminal device 100 . . In this process, the service device 200 extracts recognition metamaps having a similarity greater than or equal to a specified size to the recognition metamap for search included in the search request information from among the recognition metamaps stored in the memory, and based on the extracted recognition metamaps You can configure the recommended list with . The recognition meta map identifies the content with a designated recognition technology and a designated recognition unit for content having a certain playback time, extracts the recognition meta for the content, and displays the extracted recognition meta information in the form of a two-dimensional map. may include Accordingly, the recognition meta map may include the type of recognition technology, recognition unit information applied to the recognition technology, and recognition metas.

The network environment 10 supporting the content operation function of the present invention as described above configures and stores contents into recognition meta maps in relation to content management, and stores recognition meta maps in response to a search request requested by the terminal device 100 . By performing the comparison, it is possible to provide a faster and more efficient search function, while searching for and providing more accurate and reliable similar content.

Meanwhile, in the above description, it has been described that the terminal device 100 provides a recognition meta map for searching for content, but the present invention is not limited thereto. The terminal device 100 may input a plurality of pieces of search information (eg, thriller, comedy, romance, etc.) in relation to the search. The service device 200 may convert the plurality of pieces of search information into recognition meta for search, and search the recognition meta map including information related to the recognition meta for search. For example, when the user provides search information in thriller 50%, comedy 30%, and romance 20%, the service device 200 determines a recognition meta for search information (thriller: 01, comedy: 10) , can be determined with binary data, such as Romance:11). Then, the service device 200 searches the recognition meta map having the recognition meta value for search in a form corresponding to the search information, and configures the search data within a certain similarity range as a recommendation list to form the terminal device 100 . can be provided to

2 is a diagram showing a schematic form of a service apparatus according to an embodiment of the present invention.

Referring to FIG. 2 , the service device 200 of the present invention may include a communication circuit 210 , a memory 240 , and a processor 250 .

The communication circuit 210 may form a communication channel of the service device 200 . For example, the communication circuit 210 may establish a communication channel with the terminal device 100 in response to a request to access the terminal device 100 . The communication circuit 210 may provide a virtual page (eg, a content use screen) designated to the terminal device 100 in response to the control of the processor 250 . The communication circuit 210 may receive search request information related to content search from the terminal device 100 . The communication circuit 210 may provide a recommendation list generated in response to the search request information to the terminal device 100 . The communication circuit 210 may transmit the content selected by the request of the terminal device 100 to the terminal device 100 in a streaming method or a download method.

The memory 240 may store various data, programs, algorithms, etc. related to the operation of the service device 200 . For example, the memory 240 may include a content DB 241 and a recognition meta map DB 243 .

The content DB 241 may include at least one content. For example, the contents DB 241 may store various image contents such as movies, dramas, music videos, and documentaries. The image contents stored in the content DB 241 may be mapped to recognition meta maps stored in the recognition meta map DB 243 .

The recognition meta map DB 243 may store recognition meta maps generated from image contents stored in the content DB 241 . The recognition meta maps may include recognition metas that recognize information included in the content of the image content using at least one recognition technology and a designated recognition unit. Additionally, the recognition meta maps may further include title information of the corresponding content, time information related to content creation, content creator or actor information, and the like.

The processor 250 may transmit data or process data related to the operation of the service device 200 . For example, in connection with performing the content operation function of the present invention, the processor 250 may access the terminal device 100 , provide a content use screen, receive search request information from the terminal device 100 , and retrieve the received search information. It is possible to process the provision of a recommendation list corresponding to the request information and the provision of content according to the selection of the terminal device 100 . In particular, the processor 250 of the present invention searches for the content corresponding to the search information included in the search request information, in the form of a two-dimensional map, extracts recognition metas extracted by using a recognition technology preset in a recognition unit designated for the content of the content. A search can be performed using the recognition meta map composed of . The search request information provided by the terminal device 100 may include a recognition meta map for search, a ratio of a plurality of recognition meta for search and recognition meta for search, or at least a part of content used for search. have. The processor 250 detects at least one recognition meta map having a high similarity (or having a similarity greater than or equal to a specified size) among the recognition meta maps stored in the recognition meta map DB 243 according to the information provided in the search request information, and Based on this, you can create a recommendation list. Such a processor 250 may include a configuration as shown in FIG. 3 .

3 is a diagram illustrating an example of a processor of a service device according to an embodiment of the present invention.

Referring to FIG. 3 , the processor 250 includes a content collection unit 251 , a recognition meta extraction unit 253 , a map construction unit 255 , a map feature classification unit 257 , a compression processing unit 258 , and a search support unit. (259).

The content collection unit 251 may collect images, store the collected images, and process an original file to be recognized by at least one recognition technology. For example, the content collection unit 251 converts an image for recognizing an actor with respect to specific video content into a frame image for each required unit, or extracts only a sound source WAV file from a VOD file of a movie to recognize music. to perform basic operations for extracting meta, such as partitioning. The content collection unit 251 may collect image content from the content providing server in real time or at a predetermined period, and perform the above-described basic operation on the collected image content.

The recognition meta extraction unit 253 extracts recognition meta for each content stored in the content DB 241 based on at least one recognition technology. For example, the recognition technology may include a facial expression recognition technology. The recognition meta extraction unit 253 may recognize the expression of at least one actor in the movie content using expression recognition technology, and extract a recognition meta for recognition. For example, when it is assumed that the expression of an actor is recognized in a specific movie content, since the expression of an actor can be changed instantaneously, if it is assumed that the recognition unit is 1 frame, the recognition meta extractor 253 is 1 Metadata can be extracted once every 29.97 seconds (sec). In this case, the recognition meta extraction unit 253 may extract recognition metas including a record of 323,676 facial expression data for a 180-minute movie.

Meanwhile, the recognition meta extraction unit 253 may collect recognition meta using sound source recognition technology. For example, when a specific song is played as background music in a specific movie, the recognition meta extractor 253 may set the recognition unit of a technology for recognizing a specific song based on sound source fingerprinting to 7 seconds. In this case, when the recognition meta extractor 253 divides the sound source recognition unit into 7-second units for a 180-minute video (eg, about 10800 seconds), about 1543 records can be collected as recognition meta. The recognition meta extraction unit 253 may collect recognition metas for the same content using different recognition techniques.

The recognition meta extractor 253 may express the recognition meta using the recognized time unit information, whether or not to recognize and a recognition value. In addition, the recognition meta may be recognized within the category of data designated by the administrator in advance. That is, the recognition meta extraction unit 253 may limit the number of expressions of the recognition meta according to the control of the administrator. For example, when the administrator has previously set four expression categories, such as happiness, sadness, anger, and embarrassment, as recognizable expressions, the recognition meta extraction unit 253 sets a total of four expression categories and no recognition result. Five predefined recognition metas can be collected as a result. Similarly, when the manager designates the number of songs to be recognized in advance for sound source recognition, one recognition result from the number of recognizable songs can be extracted as a recognition meta. Therefore, the administrator can easily grasp the number of types of recognizable recognition meta.

As another example, when the recognition meta extraction unit 253 extracts music genre information with respect to the image content using sound source recognition technology, for example, 0 is no song, 1 is ballad, 2 is rock, and 3 is dance. By extracting the recognition meta, genre information about sound sources included in the video content can be collected. Meanwhile, in the above description, values such as 0 or 1, or 0 to 3 are exemplified for the recognition meta, but in order to write more detailed information, it is possible to support classification of recognition meta values such as various bits, for example 8 bits (256 kinds). .

The map construction unit 255 may construct a map as shown in FIG. 4 or 5 using the recognition metas extracted by the recognition meta extraction unit 253 . 4 illustrates an example of a recognition meta map in the form of a two-dimensional dynamic array map according to an embodiment of the present invention. In the drawing, the axis (eg, X axis) of the two-dimensional array of the recognition meta map means time information. For example, if the X and Y coordinates are (00,00), it means the 0th time unit of the recognition engine (or recognition technology). The recognition meta extraction unit for video content can be 1 frame unit, according to the administrator setting or according to the video playback time of the video content, according to the time unit of the recognition engine, such as 10 frames or 1 second, 10 seconds, 100 seconds. It can be flexibly adjusted. For example, the recognition meta extractor 253 may set the recognition unit to be longer as the image reproduction time of the video content is longer, and may set the recognition unit to be shorter as the image reproduction time is shorter.

Assuming the recognition meta extraction by the facial expression recognition engine in units of one frame, recognition metas can be expressed as a two-dimensional square matrix with X and Y lengths of 569 (569*569 = 323,761), respectively. In this way, the recognition meta extractor 253 can flexibly adjust the size of the array to fit the length of the content, and set the scale to enable simultaneous calls according to each recognition unit to dynamically expand and contract. . In this case, the recognition meta extractor 253 may read the recognized result data of the same time unit as it is through an application program (or a playback player) that inquires a map. In the case of FIG. 4, as a binary array, whether there is a recognition meta that can be used to recommend content, for example, whether a recognizable actor appeared at a specified time point during content playback, an object or a specific object preset in the content In order to express whether an actor's expression, a preset situation, or a specified sound source exists, recognition metas expressed as binary data of two values 0 and 1 are expressed in a map form.

5 illustrates another example of a recognition meta map in the form of a two-dimensional dynamic array map according to an embodiment of the present invention. For example, in order to express more data in one array, recognition metas may be expressed as multiple bits in one array as shown in FIG. 5 . As shown in FIG. 5 , the map construction unit 255 may express the recognition result such as 00, 01, 10, 11, etc. as a 2-bit status bit. The recognition meta extractor 253 may express by increasing the number of bits as the number of recognizable types of the recognition engine increases.

With respect to the management of recognizable species, the processor 250 may perform processing on the recognizable species based on the structure shown in FIG. 6 . 6 is a diagram illustrating an example of a structure for species management of a recognition meta according to an embodiment of the present invention. Referring to FIG. 6 , the service device 200 includes an application unit 650 in which an application program including a player, a search engine, and a meta utilization program is disposed, a hash unit 243a and meta result storage data 243b ( Example: a part of the recognition meta map stored in the memory 240). The application unit 650 may correspond to the processor 250 , and at least a portion of the hash unit 243a and the stored data 243b may be included in the configuration of the memory 240 . As the number of recognition types increases, the processor 250 may generate and manage more corresponding hash tables. Assuming a recognition meta expressing the division of places in the image, the processor 250 may set 00 as None, 01 as indoor, 10 as outdoor, and 11 as a place meta in which indoor and outdoor are mixed. The processor 250 may generate a hash table for the determined recognition meta values, and decode or encode each value of each array of the recognition meta map based on this. These recognition metas may be simultaneously viewed as shown in FIG. 7A . 7A is a diagram illustrating that it is possible to simultaneously inquire recognition meta maps of four types of recognition meta in a certain axis direction. For example, meta values written in the recognition meta map may be compared with each other through a designated inquiry method for recognition meta maps composed of the same recognition types. The meta value displayed by the recognition meta map can be stored/managed/utilized as a multidimensional map by matching face recognition information in the image to the same scale.

The map feature classifying unit 257 may extract features of the recognition meta maps and classify them based on the extracted features. For example, recognition metamaps may be classified based on ratio distribution information of meta values included in specific recognition metamaps, at least one recognition meta value representing specific recognition metamaps, and the like. Alternatively, the map feature classifier 257 may classify the recognition meta maps according to the type of recognition technology applied to the recognition meta maps and the size of recognition units.

As described above with reference to FIGS. 4 to 5 , the compression processing unit 258 vectorizes and stores the metadata, but may compress and store the metadata according to a designated method. For example, the compression processing unit 258 may compress and store only the difference value in the axial direction in the process of storing the recognition meta map.

The search support unit 259 may search recognition meta maps stored in the recognition meta map DB 243 according to a specified method. For example, as shown in FIG. 7B , genres of music recognized from specific movie contents a, b, and c may be expressed as recognition meta maps. 7B is a diagram illustrating examples of recognition meta maps for various movie contents. The search support unit 259 may calculate similarity by comparing meta information of recognition meta maps with each other. When image clustering is performed or image classification through deep learning is applied, the search support unit 259 can confirm that the similarity between movies a and b can be calculated to be relatively higher than that of movie c. The search support unit 259 may support a movie recommendation function based on the above-described similarity calculation result. That is, the search support unit 259 may implement and provide sophisticated recommendation logic using the recognition meta within the actual movie.

As described above, the service device 200 according to an embodiment of the present invention may provide a more sophisticated movie recommendation by using a recognition meta map, out of a simple text meta. For example, in addition to the genre meta of movie music, the service device 200 provides other attributes of recognition meta, such as classification according to place using place recognition technology, recognition meta such as time or season extracted by context meta recognition technology, etc. can provide sophisticated movie recommendations extended to . Also, the service device 200 may recommend a movie based on the recognition meta for the cast member information. For example, the service device 200 can easily compare the frequency or degree of appearance of a specific actor in a movie through the recognition meta map, so that an actor can easily and quickly search for a movie in which an actor appears to a certain degree or more in any movie, The results can be provided.

Meanwhile, the service device 200 of the present invention may provide a scale in/out function irrespective of a movie scale and a type of recognition meta. In particular, the service device 200 may increase the sound source recognition result data into frame unit result data when it is assumed that every 7 seconds is a result of processing for recognition metas with different time units, for example, in the case of sound source recognition. Also, as shown in FIG. 7 , the recognized result data of the same time unit can be read and used as it is with the same player based on the multidimensional overlapping vector result values.

8 is a diagram illustrating an example of storing a recognition meta map according to an embodiment of the present invention.

Referring to FIG. 8 , in the recognition meta storage method according to an embodiment of the present invention, in step 801 , the processor 250 of the service device 200 may input content. Alternatively, the content collection unit 251 of the processor 250 may collect the content.

In operation 803 , the processor 250 may generate a dynamic arrangement (eg, a recognition meta map) tailored to the number of recognition units and species of the recognition technology. In this operation, the processor 250 may designate the size of the array and the number of bits according to the reproduction time size of the content. For example, in consideration of the data efficiency stored in the memory 240 , the processor 250 sets the number of recognized species (or types) to be relatively small when the reproduction time of the content is relatively long, and the When the content reproduction time is relatively short, the number of the recognized species may be set to be relatively large to limit the amount of data to be extracted from one content within a specified size. Alternatively, the processor 250 may adjust the number of recognized species differently according to the level of detail of the content-related recognition meta. For example, the processor 250 increases the number of recognized species when a relatively high level of detail is requested for specific content (eg, when a high level of detail is requested from a terminal device or an advertiser device), and a relatively low level of detail is increased. If requested, the number of recognized species may be reduced.

In operation 805, the processor 250 may generate a hash table of the recognized species. In step 807 , the processor 250 may recognize and store content meta through recognition technology. That is, the processor 250 may configure the recognition meta map based on the recognition meta values generated according to the hash table, and store the recognition meta map in the memory 240 .

The processor 250 of the service device 200 may support simultaneous reading of the recognition meta maps stored in the memory 240 . For example, upon receiving search request information from the terminal device 100 , the processor 250 of the service device 200 may identify a recognition technology and a recognition unit in response to the received search request. The processor 250 may extract recognition metamaps that satisfy the condition corresponding to the search request from among the recognition metamaps to which the recognition technology and recognition unit are applied identically, and generate a recommendation list based thereon.

Alternatively, the processor 250 may check the recognition technology and the recognition unit in response to the search request, but may adjust the scale to modify the recognition metamaps to have the same scale.

Alternatively, the processor 250 may provide a plurality of results in response to a search request. For example, when a plurality of recognition technologies and recognition units are included in the search request, the processor 250 collects recognition meta maps corresponding to the plurality of recognition technologies and recognition units applied to each recognition technology, and collects the collected recognition meta maps. In the middle, the terminal device 100 may search the recognition meta map of a condition matching the search-requested content. During the search process, the processor 250 may simultaneously search conditions matching the search content among the collected recognition meta maps in parallel to improve the search speed.

The search request information may include, for example, a search recognition meta map related to search content generated by the terminal device 100 . Alternatively, the terminal device 100 may include a recognition meta map for search generated by applying a designated recognition technology and recognition unit by the service device 200 based on the provided content. Alternatively, the search request information may include at least one of values indicating a recognition species used for recognition meta extraction. For example, the search request information may include at least one of information on recognized species included in the hash table described with reference to FIG. 6 .

As described above, the content management function based on the recognition meta map according to the embodiment of the present invention has a fine recognition unit, so even if the number of records of the recognition result is greatly increased, data can be simply processed with one arrangement for each recognition engine. From the storage point of view, the storage efficiency can be greatly improved. In addition, the present invention dynamically adjusts the units of the array (eg, recognition meta map) on the same scale and reads the sequence of the array as it is when inquiring, so that the recognition meta of various units can be read simultaneously, so that the recognition meta of different time units However, the recognition result data can be increased to frame unit result data, and the same player can read the recognized result data in the same time unit as it is. In particular, the present invention can achieve various application effects through image processing because array data such as a recognition meta map can be processed like a single image. Therefore, the fact that people can extract meaning from the meta itself can maximize the significance of the media recognition meta.

In addition, the content operation function based on the recognition meta map of the present invention can classify similar content by using image processing technology by analyzing various patterns of images. For example, in the present invention, when viewing a meta image map, it is possible to classify and cluster contents having a similar shape pattern, and also to classify and manage them as related contents or related data using image-based deep learning technology. Additionally, a compression technique may be applied to the recognition metamap of the present invention to enable efficient storage.

9 is a diagram illustrating an example of a method for operating a service device according to an embodiment of the present invention.

Referring to FIG. 9 , in relation to the service device operating method of the present invention, the processor 250 of the service device 200 may acquire content in step 901 . For example, the processor 250 may acquire at least one content stored in the memory 240 .

The processor 250 may perform recognition meta extraction in step 903 . In this regard, the processor 250 may apply at least one recognition technology to the obtained content and extract recognition metas corresponding to the applied recognition technology.

In step 905 , the processor 250 may generate a recognition meta map. That is, the processor 250 may configure the recognition meta map by arranging the extracted recognition meta in the form of a two-dimensional map. The processor 250 may generate a recognition meta map for a plurality of contents stored in the memory 240 . Alternatively, when new content is received from the outside, the processor 250 may generate a new recognition meta map for the received content.

In operation 907 , the processor 250 may classify the contents based on the recognition meta maps corresponding to the contents. In this operation, the processor 250 may extract recognition meta maps having a similar map configuration by comparing the recognition meta maps with each other, and classify similar contents based thereon.

In step 909 , the processor 250 may generate classification information according to content classification and store it in the memory 240 . The classification information stored in the memory 240 may be used when receiving a content search request from the terminal device 100 , and may also be used to provide a similar content recommendation list to the terminal device 100 .

The method according to the present invention may be implemented in the form of software readable by various computer means and recorded in a computer readable recording medium. Here, the recording medium may include a program command, a data file, a data structure, etc. alone or in combination. The program instructions recorded on the recording medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. For example, the recording medium includes a magnetic medium such as a hard disk, a floppy disk, and a magnetic tape, an optical recording medium such as a compact disk read only memory (CD-ROM), a digital video disk (DVD), and a floppy disk. Magneto-Optical Media, such as a disk, and hardware devices specially configured to store and carry out program instructions, such as ROM, Random Access Memory (RAM), Flash memory, etc. do. Examples of program instructions may include high-level language codes that can be executed by a computer using an interpreter or the like as well as machine language codes such as those generated by a compiler. Such hardware devices may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

Further, implementations of the functional operations and subject matter described herein may be implemented in other types of digital electronic circuits, or in computer software, firmware, or hardware including the structures disclosed herein and structural equivalents thereof, or these It can be implemented by a combination of one or more of them. Implementations of the subject matter described herein relate to one or more computer program products, ie computer program instructions encoded on a tangible program storage medium for execution by or for controlling the operation of an apparatus according to the invention. It can be implemented as the above modules. The computer readable medium may be a machine readable storage device, a machine readable storage substrate, a memory device, a composition of matter that affects a machine readable radio wave signal, or a combination of one or more thereof.

Moreover, while this specification contains numerous specific implementation details, these should not be construed as limitations on the scope of any invention or claim, but rather on features that may be specific to particular embodiments of particular inventions. It should be understood as an explanation. Certain features that are described herein in the context of separate embodiments may be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment may also be implemented in multiple embodiments, either individually or in any suitable subcombination. Furthermore, although features operate in a particular combination and may be initially depicted as claimed as such, one or more features from a claimed combination may in some cases be excluded from the combination, the claimed combination being a sub-combination. or a variant of a sub-combination.

Likewise, although acts are depicted in the figures in a particular order, it should not be construed that all acts shown must be performed or that such acts must be performed in the specific order or sequential order shown in order to achieve desirable results. In certain cases, multitasking and parallel processing may be advantageous. Further, the separation of the various system components of the above-described embodiments should not be construed as requiring such separation in all embodiments, and the program components and systems described may generally be integrated together into a single software product or packaged into multiple software products. You have to understand that you can.

Certain embodiments of the subject matter described herein have been described. Other embodiments are within the scope of the following claims. For example, acts recited in the claims may be performed in a different order and still achieve desirable results. As an example, the processes illustrated in the accompanying drawings do not necessarily require the specific illustrated order or sequential order to achieve desirable results. In certain implementations, multitasking and parallel processing may be advantageous.

The present description sets forth the best mode of the invention, and provides examples to illustrate the invention, and to enable any person skilled in the art to make or use the invention. This written specification does not limit the present invention to the specific terms presented. Accordingly, although the present invention has been described in detail with reference to the above-described examples, those skilled in the art can make modifications, changes, and modifications to the examples without departing from the scope of the present invention.

Therefore, the scope of the present invention should not be defined by the described embodiments, but should be defined by the claims.

The present invention is applied to the field of content management.

In particular, the present invention makes it possible to easily recognize the overall characteristics or contents of the contents by configuring the contents of the reproduction frames of the contents as a recognition meta map for contents, and based on this, the contents that the user desires to search are more While it is easy and fast to search, it can increase the reliability of the search results.

10: Network environment
100: terminal device
150: network
200: service device
210: communication circuit
240: memory
243a: hash part
243b: save data
250: processor
251: content collection unit
253: recognition meta extraction unit
255: map composition part
257: map feature classification unit
258: compression processing unit
259: Search Support
650: application

Claims (10)

Memory;
a communication circuit forming a communication channel with the terminal device; and
It is functionally connected to the memory, obtains the content stored in the memory, recognizes the content of the content as a preset recognition unit from the content using at least one recognition technology, extracts recognition metas, and the extracted recognition meta A processor for generating recognition metamaps configured in the form of a map by vectorizing each content based on the
the processor
The communication circuit by extracting recognition metamaps having a similarity greater than or equal to a specified size to the recognition metamap for search included in the search request information from among the recognition metamaps stored in the memory, and generating a recommendation list based on the extracted recognition metamaps Service device, characterized in that it is set to provide to the terminal device through. According to claim 1,
the processor
The service apparatus of claim 1, wherein the content is classified based on the recognition meta maps, and classification information according to the classification is stored in the memory. According to claim 1,
the processor
and generating a recognition meta map for search by applying a recognition technology and a recognition unit designated for the content for search when receiving the content for search from the terminal device. 4. The method of claim 3,
the processor
extracting recognition type information necessary for a recognition meta map search and a distribution value of the recognition type information from among the information included in the search request information including the recognition meta map;
The service apparatus of claim 1, wherein the recognition type information is configured to detect a recognition metamap having a distribution value within a predetermined similarity to the specified distribution value. delete According to claim 1,
the processor
The service apparatus according to claim 1, wherein the number of recognition types to be applied to the recognition metamap is differently set according to the length of the reproduction time of the content. 7. The method of claim 6,
The recognition meta map includes information that two-dimensionally arranges recognition metas recognized based on the at least one recognition technology with respect to at least a portion of the content reproduction frames of the content. According to claim 1,
the processor
and collecting recognition technology and recognition unit information included in the search request information including the recognition meta map, and collecting a plurality of recognition meta maps corresponding to the recognition technology and recognition unit information. service device,
obtaining contents stored in the memory;
extracting recognition metas by recognizing the content of the content as a preset recognition unit from the content using at least one recognition technology;
generating recognition meta maps configured in the form of a map by vectorizing each content based on the extracted recognition metas;
classifying the contents based on the recognition meta maps;
storing classification information according to the classification in the memory;
extracting recognition metamaps having a similarity greater than or equal to a specified size to a search recognition metamap included in search request information from among the recognition metamaps stored in the memory; and
generating a recommendation list based on the extracted recognition meta maps and providing it to a terminal device through a communication circuit;
Service device operating method comprising a. A computer recording medium storing at least one instruction executed by a processor,
The at least one instruction is
service device,
obtaining content stored in the memory;
extracting recognition metas by recognizing the content of the content as a preset recognition unit from the content using at least one recognition technology;
generating recognition meta maps configured in a map form by vectorizing each content using the extracted recognition metas;
classifying contents based on a plurality of recognition meta maps;
storing classification information according to the classification in the memory;
extracting, from among the recognition metamaps stored in the memory, a recognition metamap for a search included in the search request information and a recognition metamap having a similarity greater than or equal to a specified size; and
generating a recommendation list based on the extracted recognition metamaps and providing it to a terminal device through a communication circuit;
Computer recording medium, characterized in that set to perform.

Images