FR3105689A1 - Analysis of multimedia content - Google Patents

Abstract

TITLE: Multimedia Content Analysis The invention relates to a method for analyzing multimedia content and an associated learning method. The analysis method is characterized in that it comprises, on an analysis device (DAN), the following steps: - obtaining (E30) a list of preferred events (EVS ') relating to multimedia content; - obtaining (E30) an analysis model (MA) of multimedia contents; - analyze (E31) a multimedia content (C) in real time using the analysis model; - generate an estimate of the presence, in the multimedia content (C), of at least one event (EV) from the list of favorite events (EVS '). fig. 3

Description

Analysis of multimedia content

Field of the invention

The invention generally relates to telecommunications, and more specifically to access to audiovisual content. It applies more particularly to user terminals and learning platforms.

Today, there are methods for triggering an action on audiovisual content, for example a recording or a change in the viewing of the content (zapping). However, these actions are generally triggered on the basis of information inserted into the video stream (also called metadata). For example, it is possible to change the broadcast channel when the theoretical broadcast time of a program is reached, on the basis of such metadata.

However, this assumes that this information is inserted in advance into the video stream, which is tedious and sometimes inefficient.

Moreover, if content is captured and broadcast in real time (for example a live television channel), it is not possible to launch actions related to its content, which is not necessarily known in advance.

US Patent No. 12/059,618 describes a personalized event notification system using real-time video analytics. In one embodiment, a live video stream is automatically monitored and characteristics of the video stream are analyzed against criteria that relate to an event previously identified by a user. The criteria relate to the characteristics of the image or audio in the video stream. If the criteria are met, the user is notified that the event has occurred and this notification can be done through messages on the video display screen (e.g. on the TV screen) or other messaging techniques (for example, SMS or e-mail). In one example, the criteria may define a particular logo and the user is notified if the analysis detects the logo in the video stream.

However, the event defined by this method remains limited to the recognition of a pattern contained in the audio or video stream. It is based on the detection of predefined characteristic elements in the video. As a result, this approach cannot estimate more complex events, such as the level of violence in certain content. In addition, it is often faulted because it cannot take into account the variations that may accompany the event. Moreover, this method is very complex to implement for several contents and several events.

The invention improves the state of the art.

To this end, it proposes a method for analyzing multimedia content, the method being characterized in that it comprises, on an analysis device, the following steps:
- obtain a list of favorite events relating to multimedia content;
- obtain a multimedia content analysis model;
- analyzing multimedia content in real time using the analysis model;
- generating an estimate of the presence, in the multimedia content, of at least one event from the list of preferred events.

Advantageously according to the invention, prior knowledge of a multimedia content is not necessary to trigger the estimation of the presence of an event in this content. The estimation of the presence of the event depends on a real-time analysis of the images and sounds that make up the content and no longer on information describing the content determined in advance, nor on a predetermined pattern to analyze in sequence.

“Multimedia content” means digital content that contains at least one image, text or sound data. The multimedia content can be received in broadcast (streaming, television channel, radio, etc.) or stored on a medium (hard disk, database, server, etc.).

“Event” means a characteristic or a set of characteristics linked to at least one of the media constituting the content; in connection with the previous examples, it could be a song, an actor, a level of violence, a face, etc.

By “analysis”, we mean a process applied to the multimedia content to deduce the presence of at least one event. The analysis can output, in connection with the previous examples, an estimate of the presence of the song, the actor, a level of violence of a certain threshold, etc. This estimate can be accompanied, for example, by a probability, or a reliability index.

By "real-time" analysis is meant an analysis of the multimedia content in association with a time base (reception time for broadcast content, reading time for stored content).

According to a first particular embodiment of the invention, a method as described above further includes the steps of:

- obtain at least one favorite event of at least one user, relating to multimedia content;

- make the multimedia content available according to the result of the estimation of the presence of the preferred event.

By "making available" we mean any action that informs the user of the availability of multimedia content: notification message, recording (on a medium accessible to him), routing of content, switching of content or channel, etc.

Thus, for example, a user will be able to automatically view a concert when his favorite song is broadcast on a channel, or available for streaming, or recorded in the local network; be automatically notified of a movie in which their favorite actor plays; being informed of the level of violence in a film, etc. This invention therefore allows the operator who manages the content of a user and the people who consume the video stream not to be dependent on the metadata provided by the television channels to provide this type of service. All the user has to do is inform this operator of the criteria that interest them for the multimedia content to be made available to the customer.

According to another particular embodiment of the invention, which may be implemented alternatively or cumulatively with the previous one, in a method as described above, the analysis model is that of a neural network and includes the optimized coefficients of the neural network, and the analysis further includes a step of:
- implement the neural network on the multimedia content, with the optimized coefficients.

Advantageously, according to this mode, a neural network is used to perform the content analysis. Such networks are known for their ability to deal with diverse and varied problems. They also make it possible to deal with unstructured problems, that is to say problems on which no information is available beforehand. The neural network can for example detect with a certain index of reliability a level of violence in the content submitted to it.

According to a variant of the first embodiment, the provision consists in giving access to the multimedia content to at least one terminal of the user.

Advantageously according to this variant, if the analysis step concludes with the presence in the analyzed content of the event that the user has designated as one of his favorites, the content is made available for the user, in particular for viewing or recording. The content could, for example, be routed (streaming, downloading, broadcasting, etc.) to the user's home gateway, to be delivered to their TV screen, or recorded on a hard drive on the local network, etc.

According to another variant of the first embodiment, which may be implemented alternatively or cumulatively with the previous one, the provision consists in notifying the presence of the content, on at least one user terminal.

Advantageously according to this variant, if the analysis step concludes with the presence in the analyzed content of the event that the user has signified as one of his favorites, a notification is transmitted to the user (for example via the home gateway for display on one of its terminals) so that it can react accordingly (download content, change channels, etc.)

According to another functional characteristic, the invention also proposes a method for learning characteristics of multimedia content, the method being characterized in that it comprises, on a learning device, the steps of:
- acquiring a plurality of multimedia content;
- associating at least one event with at least one acquired content to constitute at least one learning data pair;
- generating an analysis model from the training data pairs.

Advantageously, the invention offers a learning method on several multimedia contents likely to contain events of interest for users. This learning can be done for example by an artificial intelligence (AI) engine using supervised learning. The purpose of this learning is to identify a certain number of triggering characteristics, or events, of all the videos. Once the contents and associated events have been entered into the learning device, it generates an analysis model. The analysis model includes all the necessary elements for a later analysis of a multimedia content.

According to a particular embodiment of the invention, the learning method uses a neural network and the step of generating an analysis model consists in training the neural network by the following sub-steps:
- receiving iteratively as input a pair of training data;
- optimize the coefficients of the neural network;
- build the analysis model from the optimized coefficients.

Advantageously according to this mode, it is a neural network which is used to carry out the learning. Such networks are known for their ability to deal with diverse and varied problems. They also make it possible to deal with unstructured problems, that is to say problems on which no information is available beforehand. The analysis model resulting from the neural network notably includes the optimized coefficients of the neural network, which will be able to be provided very simply to an analysis device.

According to a variant of this embodiment, the learning method takes into account a complexity criterion in its optimization.

Advantageously according to this variant, the analysis model can be adapted to the analysis device. If the latter has limited computing power, one can in particular consider making a low-complexity model, so that it can be implemented on the device.

The invention also relates to a device for analyzing multimedia content, comprising a memory, a processor, an artificial intelligence module, a module for acquiring multimedia content, configured to:

- obtain a list of favorite events relating to multimedia content;

- obtain a multimedia content analysis model;

- analyze multimedia content in real time using the analysis model;

- generate an estimate of the presence, in the multimedia content, of at least one event from the list of preferred events.

The invention also relates to a device for learning characteristics of multimedia content comprising a memory, a processor, an artificial intelligence module, a module for acquiring multimedia content, a module for generating a learning model configured to:

- acquiring a plurality of multimedia content;

- associating at least one event with at least one acquired content to constitute at least one learning data pair;

- generate an analysis model from the training data pairs.

The invention also relates to a server or service platform comprising an analysis device as described above.

The invention also relates to a server or service platform comprising a learning device as described previously.

The invention also relates to a home gateway comprising an analysis device as described above.

The invention also relates to a system for analyzing multimedia content, comprising:

- an analysis device as described above,

- a learning device as described above,

the system being configured in such a way that the analysis device receives the learning model generated by the learning device.

The invention also relates to a computer program comprising instructions for the implementation of one of the above methods according to any of the particular embodiments described previously, when said program is executed by a processor. The method can be implemented in various ways, in particular in wired form or in software form. This program may use any programming language, and be in the form of source code, object code, or intermediate code between source code and object code, such as in partially compiled form, or in any other desirable form.

The invention also relates to a recording medium or information medium readable by a computer, and comprising instructions of a computer program as mentioned above. The recording media mentioned above can be any entity or device capable of storing the program. For example, the medium may comprise a storage means, such as a ROM, for example a CD ROM or a microelectronic circuit ROM, or else a magnetic recording means, for example a hard disk. On the other hand, the recording media may correspond to a transmissible medium such as an electrical or optical signal, which may be conveyed via an electrical or optical cable, by radio or by other means. The programs according to the invention can in particular be downloaded from an Internet-type network.

Alternatively, the recording media may correspond to an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method in question.

These devices, system and computer program have characteristics and advantages similar to those previously described in relation to the methods for analyzing content and learning content.

List of Figures

Other characteristics and advantages of the invention will appear more clearly on reading the following description of particular embodiments, given by way of simple illustrative and non-limiting examples, and the appended drawings, among which:

Figure 1 illustrates the background of one embodiment of the invention;

Figure 2 illustrates an architecture of the devices involved in a system according to one embodiment of the invention;

FIG. 3 illustrates steps of a learning and analysis method according to one embodiment of the invention.

Description of an embodiment of the invention

General principle of the invention

The general idea of the invention is to rely on a learning approach which makes it possible to train an artificial intelligence module, for example a neural network, to detect events of interest for a consumer (user) multimedia content. During the learning step, the network evolves to learn to recognize a set of events of interest in a plurality of multimedia content.

During a subsequent analysis step, an automatic detection of events of interest contained in the audio or video stream associated with new content can be carried out.

Subsequently, actions relating to this content can be triggered (for example notification, zapping or recording). For example, if a user is watching channel A, knowing that channel B is broadcasting a football match, one embodiment of the invention allows him to program an automatic zapping on channel B when a goal is scored, by signaling as an event of interest the advent of a goal. In another example, a user who has specified the appearance of a favorite actor as an event receives a notification when the actor appears in content they are not watching. According to yet another example, if a user watches channel A, knowing that channel B is broadcasting a report in which he has expressed an interest, an embodiment of the invention allows him to automatically record channel B when the report begins.

Particular embodiments of the invention.

FIG. 1 represents the general context of an embodiment of the invention, in which a telecommunications system comprises a first local area network or LAN (Local Area Network, 1) and an extended type network, or WAN (Wide Area Network , 2). According to this non-limiting example, the LAN network is a home network and the WAN network is an Internet network. More broadly, a LAN network could be a corporate network and the WAN 2 network could be of any type (cellular, GSM - Global System for Mobile Communications, UMTS - Universal Mobile Telecommunications System, Wifi - Wireless, DVB- Digital Video Broadcast, etc.) without departing from the scope of the invention.

According to this example, a WAN network management element; for example a server (5) of a telecommunications operator, located for example at the "network head", is responsible for the content learning procedure.

According to this example, a second network management element (6) (a residential or professional gateway, a hub, etc.) and terminal equipment (T1, T2) are connected to the local network 1. This is respectively according to the example of a smartphone (T2) and a television (T1). These terminals are capable of receiving multimedia content from the home gateway on the local network. This content can come from the local network or from the wide area network.

Remember that a local area network, also called a home network in the following, is a computer network that connects together, with or without wires, the terminal equipment, or more simply terminals, of a house (computers, printing peripherals, storage, connected objects, etc.), able to communicate with each other. A home network includes router equipment, also commonly called a gateway, an intermediate element ensuring the redirection, or routing, of data packets between the various terminals and networks connected to it. The user of such a network can execute a given service on a given terminal having its own characteristics (for example, viewing multimedia content C).

Multimedia content C can be, without loss of generality, from a DVB, IP, etc. type network. It is recalled that a broadcast digital network of the DVB type (acronym for Digital Video Broadcast) has an underlying infrastructure that may be supported by satellite (SAT), digital terrestrial television, or even cable, not shown. Alternatively or in addition, the terminals could also be connected to a DVB-IP type network, also called IPTV, that is to say ensuring the broadcasting of digital television or the supply of content, for example in VOD (Video on Demand) from the Internet network under the control of a service operator ensuring the quality of delivery (for example, an operator's TV, associated with a digital decoder and a home gateway from the same operator). In these two cases, the multiplexed digital contents are received by a digital decoder, also called STB (Set Top Box) via appropriate reception media (antennas, cables, ADSL, etc.) and the various programs which constitute them are demultiplexed and decoded before playback by the terminals (e.g. televisions) connected to the STB. Such a decoder, not shown, can be associated with the gateway or directly integrated into the terminal. The terminal can also be provided for accessing content streamed from the WAN or LAN. In this case, we speak of “connected” television, i.e. the terminal can be connected to the Internet network in order to provide a set of services to users. If the restitution is not carried out under the control of a service operator, we also speak in this case of OTT type content, an acronym for the English name “Over The Top”. OTT television allows a user of a terminal connected to the Internet network to render digital channels or Web content on the television. Generally, this type of "connected" terminal interfaces with the gateway in charge of the user's local network and thereby accesses a content server in the extended network. Finally, such a terminal can access the content from inside the local network, for example from a streaming device associated with a local hard disk (not shown). In the following, we refer indiscriminately to multimedia content for content received from a broadcast network (DBV, IPTV), in OTT, and/or via download (streaming, file download, etc.)

In all these cases, it is assumed that the content is analyzed by a content analysis device, which can be located on a server (5) of the operator, or on the home gateway (6), on a terminal , or on any other equipment of the remote or local network capable of carrying out such an analysis.

Beforehand, training was carried out on several contents. This learning takes place on a learning device, which is preferably located on a server (5) of the operator (which may or may not be different from the analysis server) but could also be located on the home gateway, or on any other network equipment capable of carrying out such learning.

The process according to the invention therefore proceeds in two stages, or distinct phases:

First phase: learning features and events

Initially, which corresponds to a so-called learning phase, N multimedia contents (where N is a natural number) are selected, on which to learn characteristics, or events, using a module of artificial intelligence. The learning module is able to define parameters to then allow, from any content received, to output an indication of correspondence between the content received and at least one of the events processed by the learning . For example, such a characteristic relates to the presence of a given actor in the content.

This module is typically a machine learning (ML) module. It is recalled that machine learning, or statistical learning, concerns the design, analysis, development and implementation of methods allowing a machine (in the broad sense) to evolve through a systematic process, and thus to fulfill difficult or problematic tasks by more classical algorithmic means. A possible example of machine learning is that of classification, the purpose of which is to label each piece of data by associating it with a class.

According to a preferred embodiment, a neural network (RN) is used here. During the learning stage, the network evolves to learn to recognize a set of characteristic elements that confirm the presence of an event in a content. For example, if we want to teach the machine the presence of an actor in a video sequence, we present it with films and content in which this actor appears by training it to find all the sequences where he appears. The same process can be applied for the detection of a goal in a sports sequence, the identity of the team which scored, the recognition of a place, a city, certain forms of violence in the contents, a song, etc.

The learning module outputs a "model" consisting of a set of data (software, parameters, optimized coefficients of the neural network, etc.). In the following, this model is called the “analysis model” (AM).

Second phase: content analysis

In a second phase, called analysis, when content is received (or analyzed in real time on a hard disk), the analysis program is implemented on an analysis device (DAN). This content can be new content, or content that was part of the learning.

The analysis device inherits the analysis model (MA) provided by the learning device.

According to a preferred embodiment, the analysis module uses the neural network (RN), with the analysis model and the content, possibly new, as input, intended to be viewed by one or more users of the local network if their criteria (event preferences) are met. If the analysis module detects the presence (possibly accompanied by a reliability index) of a triggering event, an action can be taken with regard to the provision of the content for one or more users of the local network. For example, a user 1 of the terminal T1 will be able to receive the content directly since one of his criteria (presence of his favorite actor) has been met, whereas a user 2 of the terminal T2 will be able to receive a notification (according to which the content C contains a musical sequence that he likes).

According to one embodiment, the analysis module is located on the home gateway 6. Advantageously in this case, the preferences of the users of the local network can be taken into account. For example, the characteristics associated with a user are recorded in a profile relating to this user. The neural network parameters corresponding to the trigger characteristics can advantageously be implemented in relation to this user.

According to another embodiment, the analysis module is located on an analysis server/platform 5 of the operator. Advantageously in this case, the analysis can benefit from a higher processing power, and is moreover centralized. A content or a notification relating to the content is sent to the home gateway in the event that one of the contents received by the platform contains an event concerning the local network of the gateway.

Figure 2 illustrates an architecture of the devices involved in a system according to one embodiment of the invention.

The learning device DAP has the conventional architecture of a computer and comprises in particular a memory M, a processing unit UT, equipped for example with a processor PROC, and controlled by the computer program PGAP stored in memory M On initialization, the code instructions of the computer program PGAP are for example loaded into a memory before being executed by the processor PROC. The processor PROC of the processing unit UT notably implements the steps of the learning program according to any one of the particular embodiments described in relation to FIGS. 1 and 2, according to the instructions of the computer program PGAP .

The device also includes:

• an artificial intelligence RN module capable of performing automatic learning;
• an MA module in charge of generating the analysis model; this module is capable, in one embodiment, of generating an analysis module of variable complexity, according to a complexity criterion (imposed by the analysis device which will implement the analysis module);
• a communication module COM1 for managing all the inputs/outputs of the device, in particular acquiring the multimedia contents, and transmitting the analysis model to an analysis device if it is separate from the learning device;
• a module for obtaining the events to be processed, EVS, for example a database manager;
• optionally a base BA to store the pairs made up of the contents and their associated events supplying the module RN.

The DAN analysis device has the classic architecture of a computer and includes in particular:

• a memory M', a processing unit UT', equipped for example with a processor PROC', and driven by the computer program PGAN stored in memory M'. On initialization, the code instructions of the analysis computer program PGAN are for example loaded into a memory before being executed by the processor PROC'.
• The processor PROC' of the processing unit UT' notably implements the steps of the analysis method according to any one of the particular embodiments described in relation to FIGS. 1 and 2, according to the instructions of the program of PGAN computer. The PGAN program is notably responsible for implementing the RN artificial intelligence module with the analysis model inherited from the learning device.
• an artificial intelligence RN module;
• an MA module in charge of obtaining and implementing the analysis model;
• a communication module for managing all the inputs/outputs of the device, in particular acquiring the multimedia content to be analyzed and the analysis model MA of the learning device if it is separate from the analysis device;
• an EVS’ module capable of obtaining the event(s) of interest for the analysis.

FIG. 3 illustrates steps of the provision method according to one embodiment of the invention.

In a first learning phase P1, prior to the real-time implementation of the neural network RN, the method comprises the following steps. The method is described to correspond to a plurality of events (actors, sports sequence including a shot on goal, handover, film including scenes of extreme violence, etc.) corresponding for example to a plurality of users (together or subset of subscribers to the operator's audiovisual service, set of local network terminal users, etc.).

In a first step E0, an initialization of the learning module, according to this example a neural network, is carried out. In the following description, it is assumed that the learning takes place on a set of events denoted EVS. Network initialization is relative to this set of events. An event in the set may have been transmitted to the learning device by the service gateways, or be created by the operator, or come from a third party, etc.

In a step E1, the method acquires a plurality (preferably, a large volume) of digital contents C0…CN, preferably but not necessarily representative of the events submitted. These contents can be chosen for example by the operator, manually or automatically.

In a step E2, the method consists, from the contents acquired in the first step E1, in associating (for example by human action) with each content, a type of event, from the set of events. The set of data corresponding to the pairs (content, event) can constitute for example a learning base BA.

In a so-called learning step E3, the method implements the actual learning of the neural network RN. The neural network, initially blank, is supplied with content to transform it until it is able to perform relevant analyzes for users. To this end, it receives iteratively or successively as input a pair of data (content, event) originating for example from the learning base BA. Note that the learning base is optional, the pairs (content, event) can be provided to the neural network iteratively without using a database.

The neural network processes the pairs (content, event) at each iteration to allow the updating of the coefficients of the neural network RN, until convergence of the algorithm is obtained. When the event recognition performance is judged to be correct, the analysis model is considered stable, so learning can be stopped. The network coefficients are said to be “frozen”, “converged” or “optimized”. These optimized coefficients, as well as all the data useful for the subsequent implementation of the neural network, are stored in the so-called "analysis module" MA which can be transmitted to analysis devices to process the contents users.

For example, the neural network training to recognize the facial features of an actor retains in its learning model MA, at the end of the learning phase, optimized coefficients and possibly elements of analysis like facial features in different angles, sizes, qualities, etc. to be able to find it in any context and any content.

It will be noted that the neural network can be any, of any known type, for example a convolutional neural network. For this, the method implements, for example, an iterative algorithm of the "gradient algorithm" type with an approach called "gradient descent" and "back-propagation", to search for the coefficients of the RN neural network allowing to achieve the best classification performance.

According to an example, the neural network RN comprises:

- one or more layers intended to filter and determine the characteristic elements of the media of the content (images, sequences of images, texts and sounds, ...); for this, it implements the application of software bricks such as:

• recognition of shapes, movements, faces, sounds, voices, etc.
• estimation of the violence of a scene
• etc

-one or more connected layers performing content classification in a limited number of events;

- a final layer using a function to estimate the probabilities of content belonging to each event. A probability between 0 and 1 can be associated with each content for an event. Such a probability can be converted into a reliability index.

According to a variant, the BA learning base and/or the analysis model can come from specialized software libraries that are already preconstituted.

According to a variant, the learning method takes into account in its optimization a complexity criterion, so as to achieve a compromise between the performance and the complexity (power, time, etc.) of the calculation required. For example, it is known to those skilled in the art that the complexity of the neural network can be controlled by adding or pruning cells. According to this example, the network can therefore be adapted to the complexity of the machine hosting the analysis device used (e.g. reduce the number of layers and/or cells and/or the type of cells if the analysis device has limited, such as a network gateway terminal, router, home computer, etc.)

At the end of this first learning phase, the process consists, in a second phase P2, called analysis, in implementing in real time the neural network RN on the analysis module. The analysis module is located according to this example in the network gateway.

According to another embodiment, it is located in a server 5 of the network operator.

In a step E30, the analysis model MA comprising in particular the optimized coefficients of the neural network, resulting from the learning step, is implemented on the analysis device. It controls the neural network RN which, based on content received by the service gateway (according to another embodiment, a server 5 of the network operator), delivers a reliable estimate of the type of event ( s) present in the content, in real time.

In a step E31, if the event has been detected in the content with a sufficiently high reliability, for example if the reliability index is greater than a given threshold, a provision action can be carried out: sending the content to a terminal (or to a gateway if the analysis module is in the network), notification, recording, etc.

By way of example, the content can be viewed by the user in a step E21.

It goes without saying that the embodiment which has been described above has been given purely as an indication and in no way limiting, and that many modifications can be easily made by those skilled in the art without departing from the scope. of the invention.

In particular, the neural network can be replaced by any artificial intelligence module capable of performing the same functions: automatic learning by decision trees, statistical methods, logistic regression, linear discriminant analysis, genetic algorithms, etc.

Claims (15)

Method for analyzing multimedia content (C), the method being characterized in that it comprises, on an analysis device (DAN), the following steps:
- obtaining (E30) a list of preferred events (EVS') relating to multimedia content;
- obtaining (E30) an analysis model (MA) of multimedia contents;
- analyzing (E31) a multimedia content (C) in real time using the analysis model;
- generating (E31) an estimate of the presence, in the multimedia content (C), of at least one event (EV) from the list of preferred events (EVS'). Method for analyzing multimedia content according to claim 1, characterized in that it further comprises the steps of:
- obtaining (E10) at least one preferred event (EVF) of at least one user, relating to a multimedia content;
- making available (E32) the multimedia content according to the result of the estimation of the presence of the preferred event. Process for analyzing multimedia content according to Claim 1, characterized in that the analysis model is that of a neural network (RN) and comprises the optimized coefficients of the neural network, and in that in that the analysis further includes a step of:
- implement the neural network on the multimedia content (C), with the optimized coefficients. Process for analyzing multimedia content according to Claim 2, characterized in that the provision consists in giving access to the multimedia content to at least one terminal (T1) of the user. Process for analyzing multimedia content according to Claim 2, characterized in that the provision consists in notifying the presence of the content, on at least one terminal (T2) of the user. Method for learning characteristics of multimedia content, the method being characterized in that it comprises, on a learning device (DAP), the steps of:
- acquiring (E1) a plurality of multimedia contents (C0…CN);
- associating (E2) at least one event (EV) with at least one acquired content (C) to constitute at least one learning data pair (EV, C);
- generating an analysis model (MA) from the training data pairs (EV, C). Method for learning characteristics of multimedia contents according to claim 6, characterized in that the learning method uses a neural network (RN) and in that the step of generating an analysis model consists in training the network of neurons (RN) by the following sub-steps:
- receiving iteratively as input a pair of training data (C, EV);
- optimize the coefficients of the neural network (RN);
- build the analysis model (MA) from the optimized coefficients. Method for learning characteristics of multimedia content according to claim 6 or 7, characterized in that the learning method takes into account in its optimization a criterion of complexity. Device (DAN) for analyzing multimedia content, comprising a memory (M'), a processor (PROC'), an artificial intelligence module (RN), a module for acquiring multimedia content (COM2), configured For:
-obtaining (E30) a list of preferred events (EVS') relating to multimedia content;
- obtaining (E30) an analysis model (MA) of multimedia contents;
- analyzing (E31) a multimedia content (C) in real time using the analysis model;
- generating an estimate of the presence, in the multimedia content (C), of at least one event (EV) from the list of preferred events (EVS'). Learning device (DAP) of multimedia content characteristics comprising a memory (M), a processor (PROC), an artificial intelligence module (RN), a module for acquiring multimedia content (COM1), a module for generating a learning model (MA) configured for:
- acquiring (E1) a plurality of multimedia contents (C0…CN);
- associating (E2) at least one event (EV) with at least one acquired content to constitute at least one learning data pair (EV, C);
- generating an analysis model (MA) from the pairs (EV, C) of learning data. Server or service platform comprising an analysis device according to claim 9. Server or service platform comprising a learning device according to claim 10. Home gateway comprising an analysis device according to claim 9. System for analyzing multimedia content, comprising:
- an analysis device according to claim 9 and
- a learning device according to claim 10,
the system being configured such that the analysis device receives the learning model generated by the learning device. Computer program capable of being implemented on a device as defined in claim 9 or 10, the program comprising code instructions which, when the program is executed by a processor, carry out the steps of the method defined according to one of claims 1 to 5 or 6 to 8.