Generating and assigning metadata

EP4457658A1Inactive Publication Date: 2024-11-06TRANSFERMEDIA PROD SERVICES GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
EP2022714549
Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2022-03-16
Publication Date
2024-11-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing methods fail to effectively generate and assign metadata for action objects in audio and visual media content, making it difficult to clearly identify and trace these objects across different production steps and their associated time values.

Method used

A method and device that process data from multiple production steps to identify and assign metadata to action objects, using an algorithm-based approach to unify names and time values, and establish links between action objects, while also comparing parameters and ontology to ensure accurate assignment and linking.

Benefits of technology

This solution enables clear identification and tracing of action objects across production steps, generating metadata that includes unified names, time values, and links, facilitating efficient comparison and matching with third-party data sets, thus improving the management and analysis of action objects in media content.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1.1
    Figure 1.1
Patent Text Reader

Abstract

The aim of the invention is to generate and assign metadata relating to handling objects which arise in audio and / or visual media contents. The handling objects are visual and / or audible objects or elements which can be perceived by a viewer and / or listener in audio and / or visual media contents and and which originate from different production steps of the audio and / or visual media contents.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Title: Generation and assignment of metadata

[0002] The present invention relates to a method, a device, a computer program and a computer program product for generating and assigning metadata of action objects of an audio and / or visual media content.

[0003] Metadata is known to be used as additional information to describe audio and / or visual media content and to specify technical and descriptive properties of an audio and / or visual media content. Examples of metadata include author, creation date, recording location, camera information, upload date, image resolution, refresh rate, aperture, focal length, artist name, music style, and sample rate.

[0004] The invention is based on the object of generating and assigning metadata about action objects that appear in audio and / or visual media content. These action objects are visual and / or audible objects or elements that can be perceived by a viewer and / or listener in audio and / or visual media content, and are not physical characteristics of the digital data of a video file or an audio file. Furthermore, the invention is based on the object of making the metadata of action objects from the various production steps of the audio and / or visual media content clearly identifiable and traceable, conceptually and / or temporally.

[0005] The present invention is described below according to the claims, listed embodiments and figures.

[0006] According to a first aspect of the present invention, a method for processing and generating metadata of action objects of an audio and / or visual media content is provided, the method comprising the following steps:

[0007] - Processing data from at least two production steps of the audio and / or visual media content, whereby the data is digitized information on the respective production step.

[0008] In a further step, data elements are identified from the data relating to the action objects used in the audio and / or visual media content to create or produce the audio and / or visual media content. According to the invention, the action objects can be people, objects, services, backgrounds, locations, or generally self-contained elements of image and / or sound that are used in the respective sequences of the audio and / or visual media content based on the action of the audio and / or visual media content. Thus, in this step, data elements relating to the action objects in the audio and / or visual media content are identified.

[0009] In a further step, data elements from at least two production steps belonging to a single action object are assigned. Data elements from the at least two different production steps are processed in such a way that a clear assignment to the corresponding and single action object is achieved. Preferably, the assignment is performed in such a way that different designations and / or specifications from the at least two different production steps for the same action object are identified algorithmically, and the data elements are assigned to one another to establish that they are data elements for the same action object.Preferably, the assignment of data elements to the same action object is based on the ontology and / or the data model, so that similar and / or synonymous designations are identified in the different production steps and assigned to the respective action object. Furthermore, the assignment of data elements to the same action object is preferably based on a comparison of parameters and / or parameter values ​​that specify a particular action object and can be assigned to the same action object, so that it can be concluded that this is the same action object.

[0010] Furthermore, a uniform designation is defined for the data elements from the at least two production steps that belong to the single action object and a first time value of the occurrence time of the single action object is assigned. According to the invention, the uniform designation for the data elements is introduced so that the action object is identified with the uniform designation across different production steps and additionally receives the time value at which the action object appeared in the audio and / or visual media content. Preferably, the first time value can be a plurality of time values ​​and / or a plurality of time windows. Furthermore, a temporal offset of the first time value of the occurrence time of the action object is identified.In this case, the temporal offset is preferably identified by comparing the first time value with another time value of the occurrence time in another production step and / or a master version or a final version of the audio and / or visual media content. Subsequently, a second time value of the occurrence time of the action object in a master version of the audio and / or visual media content is calculated, with the second time value being based on the identified offset.

[0011] Further according to the invention, metadata consisting of the uniform designation and the calculated second time value is generated in the master version for the only action object.

[0012] Preferably, according to the present invention, an object of action is an object, a person or a service.

[0013] Furthermore, according to a further aspect of the present invention, segment information is additionally processed and the single action object is assigned to a specific segment from the audio and / or visual media content and the generated metadata is additionally provided with the segment information.

[0014] Furthermore, according to a further aspect of the present invention, a link is established between the single action object and at least one further action object or further single action object, and the generated metadata is additionally provided with link information between the single action object or the further single action object and the further single action object. This produces metadata that contains link information between the respective action objects, so that the action objects are linked to one another. Preferably, action objects that represent a product and / or service are linked to an action object as a person.

[0015] body.

[0016] Further according to a further aspect of the present invention, the data element of an action object has at least one predefined parameter, wherein the predefined parameter has at least one parameter value to specify the action object. Further according to a further aspect of the present invention, the occurrence time is a perception time or perception time window of the action object in an audio and / or video content by a viewer and / or listener.

[0017] Furthermore, according to a further aspect of the present invention, the assignment is carried out by a plausibility check of existing and / or missing parameters and / or parameter values ​​and / or ontology-based and / or based on a data model and / or depending on a comparison between the parameters from at least two different production steps to the same action object.

[0018] Furthermore, according to a further aspect of the present invention, information about the presence of action objects in a further data set, in particular in an action object data set of a third-party provider for products and / or services and / or a sales platform and / or a knowledge database, is used, with a step of generating a comparison list based on the action objects of the audio and / or visual media content and the existing action objects in the further data set. In this case, a comparison is made between the data set created according to the invention for the action objects of the audio and / or visual media content with the further data set with products and / or services to identify matches and / or similarities and / or differences, so that the aforementioned comparison list can subsequently be created. Preferably, the aforementioned

[0019] Comparison based on a predefined scale is used to calculate a match rate.

[0020] Preferably according to the invention, the parameter is a placeholder for color, size,

[0021] Manufacturer information, material, purchase address, time information, ingredient,

[0022] Design name, purpose of use, identification number, price,

[0023] Description of action, location, duration of visibility, statistical information,

[0024] Time of performance, duration of performance, number of performances, size of performance or

[0025] Action object specification.

[0026] According to a further aspect of the present invention, a

[0027] A computer program product is proposed that can be loaded into the memory of an electronic computer and that has instructions that, when executed by the electronic computer, cause the electronic computer to perform the steps of the methods according to one of the preceding method claims. The electronic computer is preferably a personal computer, a smartphone, a tablet PC, a server, or a notebook.

[0028] Further preferably according to the invention and in combination with one of the described embodiments, the data set of a third party provider is data of a video platform and / or a digital online shop for products and / or services.

[0029] Further preferably according to the invention and in combination with one of the described embodiments, time values ​​for the action objects or time codes of action objects are bound to the master version or the master file of the audio and / or visual media content.

[0030] Further preferably according to the invention and in combination with one of the described embodiments, the data set created according to the invention is assigned or mapped according to a predefined standard.

[0031] Further preferably according to the invention and in combination with any of the described embodiments, the production step is a script, a pre-production, a costume design, a production design, a script continuity, a footage, a film edit, a visual effect, a color correction, a sound mix, a master.

[0032] Further preferably, according to the invention, a device for processing and generating metadata of action objects of an audio and / or visual media content is proposed, wherein the device comprises: a processing unit with at least one input for receiving and processing data from at least two production steps of the audio and / or visual media content, wherein the processing unit is configured to identify data elements from the data with reference to the action objects, wherein the device comprises an assignment unit configured to assign the data elements from at least two production steps belonging to a single action object to one another, wherein the processing unit is further configured to define a uniform designation for the data elements from the at least two production steps belonging to the single action object, wherein the device comprises an assignment unit,which is configured to assign a first time value of the occurrence time of the single action object, wherein the processing unit is further configured to identify a time offset of the first time value of the occurrence time of the action object, wherein the device further comprises a calculation unit configured to calculate a second time value of the occurrence time of the action object in a master version of the audio and / or visual media content, wherein the second time value is based on the identified offset, wherein the processing unit is further configured to generate metadata consisting of the uniform designation and the calculated second time value in the master version for the single action object.

[0033] Furthermore, it is proposed according to the invention that the processing unit is further configured to process segment information and to assign the single action object to a specific segment from the audio and / or visual media content and to additionally provide the generated metadata with the segment information.

[0034] Furthermore, it is proposed according to the invention that the device further comprises a linking unit which is configured to additionally establish a link between the single action object and at least one further single action object and to additionally provide the generated metadata with linking information between the single action object and the further single action object.

[0035] Furthermore, it is proposed according to the invention that the processing unit is further configured to process information about an action object as an object, a person, a service, a real body, an animated body or a real-animated combination of a body and / or that the processing unit is configured to process a data element, wherein the data element of an action object has at least one predefined parameter and wherein the predefined parameter has at least one parameter value in order to specify the action object.

[0036] Furthermore, it is proposed according to the invention that the processing unit is configured to process the occurrence time as a perception time or perception time window of the action object in the audio and / or video content and / or that the processing unit is configured to receive and process additional data in order to carry out the assignment ontology-based and / or data model-based.

[0037] It is further proposed according to the invention that the processing unit is set up to carry out the assignment by means of a plausibility check of existing and / or missing parameters and / or parameter values ​​between at least two production steps and / or that the processing unit is set up to receive and use information about the presence of action objects in a further data set, in particular in an action object data set of a third-party provider for products and / or services, a sales platform or a knowledge database and wherein the processing unit is further set up to generate a comparison list based on the action objects of the audio and / or visual media content and the existing action objects in the further data set.

[0038] Furthermore, it is proposed according to the invention that the processing unit is set up to process the parameter as a placeholder for color, size, manufacturer information, material, purchase address, time information, ingredient, design name, reason for use, identification number, price, action description, location information, duration of visibility, statistical information, time of occurrence, duration of occurrence, number of occurrences, size of occurrence or action object specification.

[0039] Furthermore, according to the invention, a computer program is proposed, wherein the computer program comprises instructions which, when the program is executed by a device, cause the device to operate according to one of the devices described here according to the invention.

[0040] Furthermore, the invention proposes a computer program product which can be loaded into the memory of an electronic computer and which has instructions which, when executed by the electronic computer, cause the electronic computer to carry out the steps according to one of the methods described here.

[0041] The invention is described in more detail below with reference to figures.

[0042] Figure 1 shows metadata from various production steps and the generation of new metadata according to one aspect of the present invention.

[0043] Figure 2 shows a comparison of the metadata generated according to the invention with another data set from a third party provider.

[0044] Figure 1 shows metadata of action objects of an audio and / or visual media content from various production steps DPI, DP2, DP3, and DP4. According to the invention, the metadata from the first production step DPI and the second production step DP2 are identified. Furthermore, in step Z, the metadata from DPI and DP2, which belong to a single action object, are assigned, and a first time value of the time of occurrence of the respective action objects is assigned from the metadata DPI and DP2, so that new metadata DPN is subsequently available that both assigns the action objects from DPI and DP2 to each other and also contains the respective times of occurrence of the action objects. Thus, according to the invention, the metadata DPN is generated, which unambiguously specifies the action objects with the associated time of occurrence.

[0045] Figure 2 further shows, based on the description of Figure 1, further processing of the newly generated metadata DPN using another data set E from a third-party provider and additionally performs step M to perform a comparison between the metadata DPN and E. In step M, the action objects from the production steps DPI and DP2, standardized in the metadata DPN, are compared with the other data set E, and the action objects that find a correspondence in E are generated as a new data set DPE or as a comparison list. Thus, according to the invention, the relevance of the action objects from DPI and DP2 for the data set of the third-party provider E is determined using the comparison list DPE using the metadata DPN.

[0046] Further alternatively, based on Figure 2, it can be described that the data set E of the third-party provider represents a specification for displaying the metadata DPN depending on this specification, which is defined by the data set E. Consequently, according to the aforementioned alternative, the comparison list DPE can contain data on action objects that are defined as relevant by the data set E and are thus important for the third-party provider. Further alternatively, the data set E can be used as a predefined or standardized specification for displaying or formatting the metadata DPN, so that the metadata DPN is converted into a format compliant with the standard E depending on the specified standard E and then corresponds to the respective desired representation of a third-party provider as a DPE.

[0047] Further alternatively based on Figure 2, it can be described that the data set E is a distribution-related specification which, for example, depending on the platform to be used, e.g. television, internet platform, mobile distribution, the metadata DPN is adapted depending on the data set E and is then available as a platform-compliant data set DPE.

Claims

Claims:

1. A method for processing and generating metadata of action objects of audio and / or visual media content, wherein the method comprises the following steps: - Processing data from at least two production steps of the audio and / or visual media content, - Identifying data elements from the data relating to the objects of action, - Assigning data elements from at least two production steps that belong to a single action object, - Establishing a uniform designation for the data elements from the at least two production steps that belong to the single object of action, - Assigning an initial time value to the time of appearance of the single object of action, - Identifying a time offset of the first time value of the appearance time of the object of action, - Calculating a second time value of the appearance time of the object of the action in a master version of the audio and / or visual media content, wherein the second time value is based on the identified offset, - Generating metadata consisting of the uniform name and the calculated second time value in the master version for the single action object.

2. The method according to claim 1, wherein the method additionally processes segment information and assigns the single object of action to a specific segment from the audio and / or visual media content and additionally equips the generated metadata with the segment information.

3. Method according to one of the preceding claims, wherein the method additionally establishes a link between the single action object and at least one further action object or further single action object and additionally equips the generated metadata with link information between the further action object or the further single action object and the further single action object.

4. Method according to any of the preceding claims, wherein an object of action is an object, a person, a service, a real body, an animated body or a real-animated combination of a body.

5. Method according to one of the preceding claims, wherein the data element of an action object has at least one predefined parameter and wherein the predefined parameter has at least one parameter value to specify the action object.

6. Method according to one of the preceding claims, wherein the time of occurrence is a time of perception or time window of perception of the object of action in an audio and / or video content.

7. Method according to one of the preceding claims, wherein the assignment is performed ontology-based and / or data model-based.

8. Method according to one of the preceding claims, wherein the assignment is carried out by a plausibility check of existing and / or missing parameters and / or parameter values ​​between at least two production steps.

9. A method according to any of the preceding claims, wherein the method further comprises: - Using information about the presence of action objects in another dataset, in particular in an action object dataset of a third-party provider of products and / or services, a sales platform or a knowledge base, - Generating a matching list based on the action objects of the audio and / or visual media content and the existing action objects in the further data set.

10. Method according to one of the preceding claims, wherein the parameter is a placeholder for color, size, manufacturer's information, material, reference address, time specification, ingredient, design name, occasion of use, identification number, price, description of action, location, duration of visibility, statistical information, time of appearance, duration of appearance, number of appearances, size of appearance or specification of the object of action.

11. Device for processing and generating metadata of action objects of audio and / or visual media content, wherein the device comprises: - a processing unit with at least one input for receiving and processing data from at least two production steps of the audio and / or visual media content, wherein the processing unit is configured to identify data elements from the data with reference to the action objects, wherein the device has an assignment unit configured to assign the data elements from at least two production steps belonging to a single action object to each other, wherein the processing unit is further configured to establish a uniform designation for the data elements from the at least two production steps belonging to the single action object, wherein the device has an allocation unit configured to assign a first time value of the occurrence time of the single action object,wherein the processing unit is further configured to identify a time offset of the first time value of the occurrence time of the action object, wherein the device further comprises a calculation unit configured to calculate a second time value of the occurrence time of the action object in a master version of the audio and / or visual media content, wherein the second time value is based on the identified offset, wherein the processing unit is further configured to generate metadata consisting of the unified label and the calculated second time value in the master version for the single action object.

12. Device according to claim 11, wherein the processing unit is further configured to process segment information and to assign the single object of action to a specific segment from the audio and / or visual media content and to additionally equip the generated metadata with the segment information.

13. Device according to claim 11 or 12, wherein the device further comprises a linking unit which is configured to additionally establish a link between the single action object and at least one further single action object and to additionally equip the generated metadata with linking information between the single action object and the further single action object.

14. Device according to claim 11, 12 or 13, the processing unit is further configured to process information about an object of action as an object, a person, a service, a real body, an animated body or a real-animated combination of a body and / or the processing unit is configured to process a data element, wherein the data element of an object of action has at least one predefined parameter and wherein the predefined parameter has at least one parameter value to specify the object of action and / or that the processing unit is configured to process the time of appearance as a time of perception or time window of perception of the object of action in the audio and / or video content and / or that the processing unit is configured to receive and process additional data.to perform the mapping ontology-based and / or data model-based and / or that the processing unit is configured to perform the mapping by means of a plausibility check of existing and / or missing parameters and / or parameter values ​​between at least two production steps and / or that the processing unit is configured to receive and utilize information about the presence of action objects in another data set, in particular in an action object data set of a third-party provider of products and / or services, a sales platform or a knowledge base and wherein the processing unit is further configured to generate a matching list based on the action objects of the audio and / or visual media content and the existing action objects in the other data set and / or that the processing unit is configured to process the parameter as a placeholder for color, size,Manufacturer's information, material, supplier's address, date, ingredient, design name, occasion of use, identification number, price, description of action, location, duration of visibility, statistical information, time of appearance, duration of appearance, number of appearances, size of appearance, or specification of the object of action.

15. Computer program product which can be loaded into the memory of an electronic computer and which includes instructions which, when executed by the electronic computer, cause the electronic computer to perform the steps of the methods according to any one of the preceding method claims 1 to 10.