CN113961898A - Detection method, device and equipment for anchor in live broadcast room and storage medium - Google Patents
Detection method, device and equipment for anchor in live broadcast room and storage medium Download PDFInfo
- Publication number
- CN113961898A CN113961898A CN202111268834.7A CN202111268834A CN113961898A CN 113961898 A CN113961898 A CN 113961898A CN 202111268834 A CN202111268834 A CN 202111268834A CN 113961898 A CN113961898 A CN 113961898A
- Authority
- CN
- China
- Prior art keywords
- current
- face information
- live broadcast
- broadcast room
- real
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims description 21
- 238000000034 method Methods 0.000 claims abstract description 37
- 238000000605 extraction Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000013473 artificial intelligence Methods 0.000 abstract description 3
- 239000013598 vector Substances 0.000 description 18
- 238000010586 diagram Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 9
- 238000012552 review Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 230000006399 behavior Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/30—Authentication, i.e. establishing the identity or authorisation of security principals
- G06F21/31—User authentication
- G06F21/32—User authentication using biometric data, e.g. fingerprints, iris scans or voiceprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Security & Cryptography (AREA)
- Computer Hardware Design (AREA)
- Collating Specific Patterns (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The disclosure provides a method, a device, equipment and a storage medium for detecting a main broadcast in a live broadcast room, and relates to the field of artificial intelligence, in particular to the field of computer vision. The specific implementation scheme is as follows: the method comprises the steps of obtaining current face information in a current live broadcast room based on video stream data of the current live broadcast room, matching the current face information with pre-stored real-name authentication anchor face information and pre-stored pilot face information of the current live broadcast room to obtain a matching result, determining whether the current live broadcast room contains a real-name authentication anchor based on the matching result, obtaining continuous duration of the current live broadcast room not containing the real-name authentication anchor, and determining whether illegal multicast generation conditions occur in the current live broadcast room based on the continuous duration. In the embodiment of the disclosure, the violation condition of the alternate broadcasting of other people is automatically detected by using the electronic equipment, all live broadcasting can be audited, the auditing coverage is improved, all live broadcasting rooms are not required to be screened manually, the auditing efficiency is improved, and the auditing cost is reduced.
Description
Technical Field
The present disclosure relates to the field of artificial intelligence technologies, and in particular, to a method, an apparatus, a device, and a storage medium for detecting a live broadcast in a live broadcast room in the field of computer vision technologies.
Background
At present, the situation of alternative broadcasting of other people occurs in the live video, and the situation belongs to violation behaviors, so that the situation of alternative broadcasting of other people needs to be detected.
Disclosure of Invention
The disclosure provides a method, a device, equipment and a storage medium for detecting a live broadcast room anchor for detecting the condition of others on behalf of a broadcast.
According to an aspect of the present disclosure, there is provided a method for detecting a live broadcast of a live broadcast room, including:
acquiring current face information in a current live broadcast room based on video stream data of the current live broadcast room;
performing face matching on the current face information, pre-stored face information of a real-name authentication anchor broadcast of a current live broadcast room and pilot broadcast face information to obtain a current face matching result;
determining whether the current live broadcast room contains the real-name authentication anchor broadcast or not based on the current face matching result;
and obtaining the continuous time length of the current live broadcast room, which does not contain the real-name authentication anchor broadcast, and determining whether the current live broadcast room has the illegal alternate broadcast condition or not based on the continuous time length.
According to another aspect of the present disclosure, there is provided a detection apparatus of a live broadcast room anchor, including:
the system comprises a current face information acquisition module, a face information acquisition module and a face information acquisition module, wherein the current face information acquisition module is used for acquiring current face information in a current live broadcast room based on video stream data of the current live broadcast room;
the face matching result acquisition module is used for carrying out face matching on the current face information, the pre-stored face information of the real-name authentication anchor broadcast of the current live broadcast room and the pilot broadcast face information to obtain a current face matching result;
the real-name authentication anchor confirming module is used for determining whether the real-name authentication anchor is contained in the current live broadcast room or not based on the current face matching result;
and the illegal multicast generation judging module is used for obtaining the continuous time length of the current live broadcast room, which does not contain the real-name authentication anchor broadcast, and determining whether the illegal multicast generation condition occurs in the current live broadcast room or not based on the continuous time length.
According to another aspect of the embodiments of the present disclosure, there is provided an electronic device including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any of the above described live room anchor detection methods.
According to another aspect of embodiments of the present disclosure, there is provided a non-transitory computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the method for detecting a live-cast anchor according to any one of the above.
According to another aspect of embodiments of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements a method of detecting a live room anchor according to any of the above.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
fig. 1 is a schematic diagram of a first embodiment of a method of detection of a live room anchor provided in accordance with the present disclosure;
fig. 2 is a schematic diagram of a second embodiment of a method of detecting a live room anchor provided in accordance with the present disclosure;
fig. 3 is a schematic diagram of a third embodiment of a method of detecting a live room anchor provided in accordance with the present disclosure;
fig. 4 is a schematic diagram of a fourth embodiment of a detection method of a live room anchor provided in accordance with the present disclosure;
fig. 5 is a schematic flow chart of an implementation of a method for detecting a live room anchor according to the present disclosure;
fig. 6 is a schematic diagram of a first embodiment of a detection apparatus of a live room anchor provided in accordance with the present disclosure;
fig. 7 is a block diagram of an electronic device for implementing a method of detecting a live room anchor of an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
With the upgrading of infrastructure of 5G popularization and the like, video becomes one of main carriers of information communication, and live video broadcast also has been widely deepened into industries such as e-commerce and e-competition. The violation detection in live video is strongly required, most of the current methods adopt reporting, human review and the like, the coverage is limited, and the cost is high.
Therefore, in order to detect the behavior of illegal multicast in the live broadcast room, the present disclosure provides a method, an apparatus, a device and a storage medium for detecting a live broadcast anchor in the live broadcast room. First, a method for detecting a main broadcast in a live broadcast room according to the present disclosure will be described below.
Referring to fig. 1, fig. 1 is a schematic diagram of a first embodiment of a detection method for a live broadcast anchor provided in an embodiment of the present disclosure, which may specifically include the following steps:
step S110, acquiring current face information in a current live broadcast room based on video stream data of the current live broadcast room;
step S120, carrying out face matching on the current face information, pre-stored face information of a real-name authentication anchor in a current live broadcast room and pilot broadcast face information to obtain a current face matching result;
step S130, determining whether the current live broadcast room contains the real-name authentication anchor broadcast or not based on the current face matching result;
step S140, obtaining the continuous time length of the current live broadcast room, wherein the real-name authentication anchor broadcast is not included in the current live broadcast room, and determining whether the current live broadcast room has the illegal alternate broadcast condition or not based on the continuous time length.
The method for detecting the anchor of the live broadcast room includes the steps of firstly obtaining current face information in the current live broadcast room based on video stream data of the current live broadcast room, then matching the current face information with pre-stored face information of a live-name authentication anchor and pre-stored face information of a pilot broadcast in the current live broadcast room to obtain a matching result, determining whether the current live broadcast room contains the live-name authentication anchor based on the obtained matching result, then obtaining continuous duration of the current live broadcast room without the live-name authentication anchor, and determining whether an illegal alternative broadcast situation occurs in the current live broadcast room based on the obtained continuous duration. In the embodiment of the disclosure, the current face information is matched with the pre-stored real-name authentication anchor face information and the pre-broadcast face information of the current live broadcast room by using the electronic equipment, so that the illegal conditions of the alternate broadcast of other people are automatically detected, all live broadcasts can be audited, the auditing coverage is improved, a large number of live broadcast rooms are not required to be screened manually, the auditing efficiency is improved, and the auditing cost is reduced.
In the embodiment of the present disclosure, current face information in each live broadcast room in which live broadcast is performed may be acquired.
In general, one or more anchor broadcasts are live in a live broadcast room. Therefore, in the embodiment of the present disclosure, the number of the acquired current face information may be one, or may be multiple.
In an embodiment of the present disclosure, referring to fig. 2, the step S110 in fig. 1 may be subdivided into:
and step S111, acquiring a live video data stream.
In the embodiment of the disclosure, the implementation can be realized through the electronic device provided with the cloud. Each live broadcast room can transmit live broadcast video data streams to the cloud in real time, and therefore the cloud electronic equipment can acquire current face information based on the live broadcast video data streams.
As a specific implementation manner of the embodiment of the present disclosure, the live video data stream may be transmitted in Real Time based on a Real Time Messaging Protocol (RTMP). Of course, other transport protocols may be used for live video streaming, which is not specifically limited in this disclosure.
And step S112, performing frame extraction on the live video data stream according to a preset time interval to obtain a frame sequence.
In the embodiment of the disclosure, after the cloud acquires the live video data streams of the live broadcasting rooms, the live video data streams acquired in real time can be framed according to the preset framing rule. As a specific implementation manner of the embodiment of the present disclosure, the preset frame extraction rule may specifically be a preset frame extraction time interval. For example, it may be set to decimate one frame every 30 seconds. Aiming at a live broadcast room, the extracted video frames are arranged according to the time sequence, and the frame sequence can be formed. As a specific implementation manner of the embodiment of the present disclosure, a frame sequence number may be set to each video frame in the frame sequence in a time sequence, and each video frame in the frame sequence may include a timestamp of the frame.
In the embodiment of the present disclosure, the frame extracting time interval may be preset based on an actual application scenario, for example, may be 30s, and certainly, may also be 15s, 60s, and the like, and is not specifically limited in the present disclosure.
And step S113, extracting the face information from each frame image in the frame sequence, and sequentially using the face information as the current face information according to the frame extraction sequence.
In the embodiment of the present disclosure, for each video frame image of the frame sequence, a CNN (Convolutional Neural Networks) may be used to extract a face feature vector in the video frame image as face information in the video frame image. As described above, there may be more than one anchor in the live broadcast room, and therefore, there may be more face information extracted from one video frame image.
After face feature vectors are extracted for each video frame image in the frame sequence, each extracted face information can be sequentially used as the current face information in the live broadcast room.
Therefore, in the embodiment of the disclosure, the face information can be extracted for each live broadcast room in live broadcast, and the occurrence probability of false detection and missed detection in the live broadcast room is reduced, so that a higher auditing coverage rate of illegal broadcast agency behaviors can be achieved.
Generally, in the field of live video, when a host registers a platform account, real-name authentication is required, and in the real-name authentication process, face data of the host is generally collected and confirmed to ensure reliability of identity information of the host. Therefore, in the embodiment of the present disclosure, each data collected in the real-name authentication process may be stored, and as a specific implementation manner, the face feature vector of the anchor face information acquired in the real-name authentication process may be extracted and stored.
In addition, a pilot broadcast is generally required to be performed before a live broadcast authority is formally acquired by a main broadcast registration platform account, and therefore, in the embodiment of the present disclosure, the main broadcast face information (e.g., face feature vectors) acquired in the pilot broadcast process may be stored for each live broadcast room.
Therefore, after the current face information in each live broadcast room is obtained, the obtained current face information can be matched with the pre-stored live broadcast room real-name authentication anchor face information and the face information obtained in the pilot broadcast process, so as to obtain a matching result of whether the current face information and the live broadcast room real-name authentication anchor face information are the same person or not.
As an embodiment of the present disclosure, as shown in fig. 2, step S120 in fig. 1 may specifically be:
and step S121, carrying out face matching on the current face information, the face information of the real-name authentication anchor in the current live broadcast room, pilot broadcast face information and at least one piece of stored historical live broadcast face information, and obtaining a current face matching result.
For each live broadcast room, the anchor of live broadcast room real-name authentication may have been normally live broadcast before, and therefore, in the embodiment of the present disclosure, the face information (e.g., the face feature vector) extracted from the live broadcast data before the anchor may be stored as the historical live broadcast face information in the historical face database to be matched with the current face information.
Meanwhile, as described above, in the embodiment of the present disclosure, the face information may be extracted from the video frame image in the frame sequence in real time, and therefore, after matching is performed on the current face information in each frame, the current face information determined to be the face information of the real-name authentication anchor may also be stored in the historical live face database as the historical live face information.
Specifically, in an embodiment of the present disclosure, before the step S121, the method may further include:
and storing the current face information as candidate historical live face information into a historical face database in real time.
In the embodiment of the present disclosure, after extracting the face information in each live video frame image, the face information may be cached in the historical face database in real time.
Correspondingly, under the condition that the current face matching result is determined to be successful in matching, the candidate historical live broadcast face information is used as confirmed historical live broadcast face information and is kept in the historical face database; and under the condition that the current face matching result is determined to be unsuccessful, deleting the candidate historical live face information from the historical face database.
That is, in the embodiment of the present disclosure, the current face information determined to be the real-name authentication anchor is stored in the historical face database to be matched with the current face information.
Therefore, in the embodiment of the disclosure, the face information of the real-name authentication anchor in the current live broadcast room, the pilot broadcast face information, and the stored at least one historical live broadcast face information can be subjected to face matching with the current face information, and the matching result is more reliable due to the increase of the matched reference data.
In an embodiment of the present disclosure, based on fig. 1, as shown in fig. 3, the step S121 may be specifically subdivided into the following steps:
step S1211, respectively calculating similarity between the current face information and pre-stored real-name authentication face information, pre-broadcast face information, and historical live broadcast face information of the live broadcast room.
As a specific implementation manner of the embodiment of the present disclosure, a ResNet50 network structure may be adopted, the extracted current face information is input into the ResNet50 network, data in a first Full Connected (FC) layer is taken as face feature information, the feature information is subjected to L2 normalization, and then is mapped into a 128-dimensional vector, cosine distances between the vector and the pre-stored real-name authenticated face feature vector, pre-broadcast face feature vector and each piece of historical live broadcast face information (face feature vector) in the live broadcast are respectively calculated, and similarity scores between the current face information and the real-name authenticated face information, the pre-broadcast face information and each piece of historical live broadcast face information are obtained based on each cosine distance (which may be respectively denoted as a, b and c). The similarity score may be a cosine value of an angle between vectors, the more the cosine value approaches 1, the more similar the vectors are.
As described above, it is possible to extract face information of a plurality of anchor persons for one video frame image in a live video data stream. Therefore, as a specific implementation manner of the embodiment of the present disclosure, after cosine similarity between each current face feature vector in a video frame image and the real-name authentication face feature vector, the pre-broadcast face feature vector, and each historical live broadcast face information (face feature vector) in the live broadcast room is obtained through calculation, the face information with the highest similarity can be used as the face information of the target anchor, and subsequently, it is only necessary to judge whether the target anchor is the real-name authentication anchor in the live broadcast room.
As shown in fig. 3, in step S1212, it is determined whether each similarity exceeds a preset similarity threshold; if the similarity exceeds the preset similarity threshold, executing step S1213; if at least one similarity value does not exceed the preset similarity threshold, performing step S1214;
step S1213, determining the current face matching result as successful matching;
step S1214, determining that the current face matching result is unsuccessful in matching.
In the embodiment of the present disclosure, a similarity threshold may be preset, and two face feature vectors exceeding the similarity threshold may be regarded as the same face, for example, the similarity threshold may be 0.95, and of course, the similarity threshold may also be other values, and may be specifically set by a developer.
As described above, for a current face information, only when the similarity of the three items a, b, and c exceeds the similarity threshold, it can be determined that the matching is successful, that is, it is determined that the anchor corresponding to the current face information is a real-name authentication anchor. Otherwise, as long as the similarity not exceeding the preset similarity threshold exists in the three similarities a, b and c, it can be determined that the anchor corresponding to the current face information is not the live-name authentication anchor in the live broadcast room.
Accordingly, as shown in fig. 3, step S130 in fig. 1 can be further detailed as follows:
step S131, if the matching is successful, determining that the current live broadcast room comprises the real-name authentication anchor;
step S132, if the matching is unsuccessful, determining that the current live broadcast room does not contain the real-name authentication anchor broadcast.
Therefore, in the embodiment of the disclosure, only if each similarity exceeds the preset similarity threshold, it is determined that the anchor corresponding to the current face information is the real-name authentication anchor, so that the reliability of the determination result is increased, the auditing accuracy is improved, and the occurrence of false detection is reduced.
In one embodiment of the present disclosure, as shown in fig. 4, step S140 in fig. 1 may be subdivided into:
and step S141, recording the time stamp of the video frame image corresponding to the current face information under the condition that the current live broadcast room does not contain the real-name authentication anchor broadcast.
Generally, each video frame in the live video stream data transmitted to the cloud end is provided with a timestamp, and the timestamp indicates the occurrence time of the video frame in the live video.
Therefore, in the embodiment of the present disclosure, after the matching result is obtained for one current face information, if it is determined that the live broadcast room does not include the real-name authentication anchor (the matching result is unsuccessful), the determination result and the timestamp of the corresponding video frame image may be cached to perform subsequent determination.
For example: the frame sequence number and the timestamp of the current live broadcast room, which do not contain the real-name authentication anchor, can be recorded in a list. As shown in table one, the sequence numbers and corresponding timestamps of the frame sequence with a decimation interval of 60s are shown in table one:
of course, if it is determined that the live broadcast room includes the real-name authentication anchor (matching is successful), the determination result and the corresponding timestamp may be cached. In the embodiment of the present disclosure, the video frame information whose matching result is successful and the video frame information whose matching result is unsuccessful may be respectively stored in a list, for example, the frame information whose matching result is unsuccessful may be stored using the first table, and the frame information whose matching result is successful may be stored using another list.
And step S142, acquiring the continuous time length of the current live broadcast room without the real-name authentication anchor based on the timestamp of the current record and the timestamp of the historical record.
In the embodiment of the present disclosure, after it is determined that the live broadcast room does not include the live broadcast room real-name authentication anchor, the continuous duration that the live broadcast room does not include the real-name authentication anchor can be obtained based on the timestamp of the current record and the timestamp of the history record, that is, the determination result of whether the live broadcast room includes the real-name authentication anchor obtained based on the current face information is integrated with each result cached in step S141.
For example, based on the above table one, if the frame sequence number of the video frame corresponding to the current face information is 1015, the timestamp is 00:30:01, and it can be seen from the table one that there are 14 (the frame sequence number is 1000 to 1014) consecutive frames before the frame, as described above, the frame sequence number and the timestamp of the real-name authentication anchor in the current live broadcasting room are stored in the table one, then the continuous duration of the real-name authentication anchor in the current live broadcasting room is 00:15:01 to 00:30:01, that is, 15 minutes and 01 seconds.
Step S143, judging whether the continuous time length exceeds a preset time length threshold value; if so, determining that the illegal alternate broadcasting situation occurs in the current direct broadcasting room; and if not, determining that the illegal multicast condition does not occur in the current live broadcast room.
In the embodiment of the present disclosure, a duration threshold (T) may be preset, and if the continuous duration of the real-name authentication anchor broadcast is not included in the live broadcast room and exceeds the duration threshold, it may be determined that an illegal situation of another alternative broadcast occurs in the live broadcast room. In a specific example, the preset time threshold may be 15 minutes. For the example in step S142, if the set duration threshold is 15 minutes, and the continuous duration of the current live broadcast room that does not include the live broadcast room real-name authentication anchor exceeds 15 minutes, it may be determined that the illegal multicast substitution condition occurs in the live broadcast room in the example. Of course, the duration threshold may be set by the live review related personnel.
Accordingly, as an implementation manner of the embodiment of the present disclosure, when the obtained determination result is integrated with the results cached before in the step S142, the obtained determination result may be integrated with the cached result of the recent period of time, and the period of time may be greater than the duration threshold. For example, if the preset time threshold is 15 minutes, the currently obtained determination result may be integrated with the cached result within half an hour (30 minutes) when the integration is performed.
When the live broadcast room does not contain the real-name authentication anchor broadcast of the live broadcast room based on the current face information, live broadcast is carried out within the backtracking time T (preset time length threshold), and the live broadcast anchor broadcast does not contain the real-name authentication anchor broadcast, the illegal alternate broadcast in the live broadcast room can be judged. Otherwise, the illegal situation of others' generation broadcasting does not occur in the live broadcasting room, and the judgment result of the real-name authentication anchor broadcasting in the live broadcasting room is added into the cache correspondingly.
As an implementation manner of the embodiment of the present disclosure, if it is determined that a violation condition of alternative broadcasting of another person occurs in the live broadcasting room, the live broadcasting can be audited by the person, and the live broadcasting room is audited manually in more detail, so that the auditing accuracy is further improved.
In the embodiment of the disclosure, after the judgment result of whether the live broadcast room contains the live broadcast room real-name authentication anchor broadcast or not is obtained based on the current face information each time, the judgment result is cached, whether the violation condition of alternative broadcast of other people occurs in the live broadcast room or not can be judged based on the current judgment result and the cached results, the live broadcast conditions of all the live broadcast rooms do not need to be audited manually, and the auditing efficiency is improved.
As shown in fig. 5, fig. 5 is a specific execution flow diagram of a detection method for a live broadcast anchor in a live broadcast room according to an embodiment of the present disclosure, and mainly includes the following steps:
firstly, frame extraction is carried out on the direct-broadcasting video data stream according to a preset time interval, and the extracted video frames are arranged according to a time sequence to obtain a frame sequence.
Secondly, extracting face information (face characteristic vector) aiming at each video frame image in the frame sequence, and taking the extracted face information as the current face information in sequence.
And thirdly, storing the face information extracted from each video frame image in the frame sequence in a database in real time, and caching the face information serving as historical face data into a historical face database.
In this disclosure, the historical face data in the historical face database is bound to the live broadcast room real-name authentication anchor, and may include current face information obtained by comparison and considered as the real-name authentication anchor, and may also include historical live broadcast face data considered as the live broadcast room authentication anchor, where the historical live broadcast face data may be: one piece of face information is kept every week, and the recent one-month anchor face data is stored, so that the labor consumption and the waste of storage space are reduced.
Comparing the current face information with pre-stored real-name authentication data, pilot broadcast extraction data and the historical face data, calculating all similarities, judging whether the current face information is consistent with the real-name authentication anchor in the live broadcast room or not based on all the similarities, and obtaining a comparison result of whether the real-name authentication anchor is contained in the live broadcast room or not.
If the current face information contains face information of a plurality of anchor, after calculating each similarity, selecting the face information with the highest similarity as the anchor face information for subsequent comparison, and clearing other non-anchor face information cached in the step (c).
And fifthly, returning the comparison result, and caching the comparison result and the timestamp of the corresponding video frame.
Integrating the cached judgment results, acquiring the continuous time length of the live broadcast room without the live broadcast room real-name authentication anchor, and judging whether the violation condition of alternate broadcasting of other people occurs in the live broadcast room based on the continuous time length. And if the illegal situation of the alternate broadcasting of other people occurs in the live broadcasting room, the live broadcasting is accessed to a human for review, and the further review is carried out manually.
Therefore, compared with the prior art, other users are encouraged to report other people's illegal acts by means of setting rewards and the like, and after the reports, the users are accessed to manual review to judge whether novel violations possibly caused by other people's illegal acts are generated, so that the review cost is increased; or directly carry out manual review to the live broadcast content to judge whether the illegal behavior of the alternative broadcast of other people causes limited coverage, low review efficiency and high review cost.
The embodiment of the present disclosure further provides a detection apparatus for a live broadcast room anchor, as shown in fig. 6, the detection apparatus may include:
a current face information obtaining module 610, configured to obtain current face information in a current live broadcast room based on video stream data in the current live broadcast room;
a face matching result obtaining module 620, configured to perform face matching on the current face information and pre-stored face information of a real-name authentication anchor broadcast and pre-stored face information of a current live broadcast room, so as to obtain a current face matching result;
a real-name authentication anchor confirming module 630, configured to determine whether the current live broadcast room includes the real-name authentication anchor based on a current face matching result;
and the illegal multicast generation judging module 640 is configured to obtain a continuous time length when the current live broadcast room does not include the real-name authentication anchor broadcast, and determine whether the illegal multicast generation condition occurs in the current live broadcast room based on the continuous time length.
The detection device for the anchor in the live broadcast room, provided by the embodiment of the present disclosure, first obtains current face information in the current live broadcast room based on video stream data of the current live broadcast room, then matches the current face information with pre-stored face information of a live-name authentication anchor and pre-stored face information of a pilot broadcast in the current live broadcast room to obtain a matching result, determines whether the current live broadcast room contains the live-name authentication anchor based on the obtained matching result, then obtains a continuous duration that the current live broadcast room does not contain the live-name authentication anchor, and determines whether an illegal alternative broadcast situation occurs in the current live broadcast room based on the obtained continuous duration. In the embodiment of the disclosure, the current face information is matched with the pre-stored real-name authentication anchor face information and the pre-broadcast face information of the current live broadcast room by using the electronic equipment, so that the illegal conditions of the alternate broadcast of other people are automatically detected, the auditing coverage is improved, a large number of live broadcast rooms are not required to be screened manually, the auditing efficiency is improved, and the auditing cost is reduced.
In an embodiment of the present disclosure, the face matching result obtaining module 620 may be configured to perform face matching on the current face information, pre-stored face information of a real-name authentication anchor in a current live broadcast room, pre-broadcast face information, and at least one stored historical live broadcast face information, so as to obtain a current face matching result.
In an embodiment of the present disclosure, the face matching result obtaining module 620 may be specifically configured to respectively calculate each similarity between the current face information and pre-stored real-name authentication face information, pre-broadcast face information, and each historical live broadcast face information of the live broadcast room;
if all the similarity exceeds a preset similarity threshold, determining that the current face matching result is successful; and if at least one similarity value does not exceed the preset similarity threshold, determining that the current face matching result is unsuccessful in matching.
Correspondingly, the real-name authentication anchor confirming module 630 may be configured to determine that the current live broadcast room includes the real-name authentication anchor if the matching is successful; otherwise, determining that the current live broadcast room does not contain the real-name authentication anchor.
In an embodiment of the present disclosure, the current face information obtaining module 610 may be configured to obtain a live video data stream;
performing frame extraction on the live video data stream according to a preset time interval to obtain a frame sequence;
and extracting face information from each frame image in the frame sequence, and sequentially using the face information as the current face information according to the frame extraction sequence.
In an embodiment of the present disclosure, the illegal substituted broadcasting judging module 640 may be configured to record a timestamp of a video frame image corresponding to the current face information when it is determined that the current live broadcasting room does not include a real-name authentication anchor;
acquiring continuous time length of a current live broadcast room, which does not contain a real-name authentication anchor, based on a timestamp of a current record and a timestamp of a historical record;
judging whether the continuous time length exceeds a preset time length threshold value or not; if so, determining that the illegal alternate broadcasting situation occurs in the current direct broadcasting room; and if not, determining that the illegal multicast condition does not occur in the current live broadcast room.
In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the personal information of the related user are all in accordance with the regulations of related laws and regulations and do not violate the good customs of the public order.
The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.
FIG. 7 illustrates a schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 7, the device 700 comprises a computing unit 701, which may perform various suitable actions and processes according to a computer program stored in a Read Only Memory (ROM)702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
Various components in the device 700 are connected to the I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, or the like; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.
It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.
The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.
Claims (13)
1. A method for detecting a live broadcast of a live broadcast room comprises the following steps:
acquiring current face information in a current live broadcast room based on video stream data of the current live broadcast room;
performing face matching on the current face information, pre-stored face information of a real-name authentication anchor broadcast of a current live broadcast room and pilot broadcast face information to obtain a current face matching result;
determining whether the current live broadcast room contains the real-name authentication anchor broadcast or not based on the current face matching result;
and obtaining the continuous time length of the current live broadcast room, which does not contain the real-name authentication anchor broadcast, and determining whether the current live broadcast room has the illegal alternate broadcast condition or not based on the continuous time length.
2. The method of claim 1, wherein,
the step of carrying out face matching on the current face information and the face information of the real-name authentication anchor broadcast and the pilot broadcast face information of the current live broadcast room stored in advance to obtain a current face matching result comprises the following steps:
and performing face matching on the current face information, the face information of the real-name authentication anchor broadcast of the current live broadcast room, pilot broadcast face information and at least one piece of stored historical live broadcast face information to obtain a current face matching result.
3. The method of claim 2, wherein,
the step of carrying out face matching on the current face information, pre-stored real-name authentication face information of the live broadcast room, pilot broadcast face information and at least one stored historical live broadcast face information to obtain a current face matching result comprises the following steps:
respectively calculating each similarity between the current face information and pre-stored real-name authentication face information, pilot broadcast face information and each historical live broadcast face information of the live broadcast room;
if all the similarity exceeds a preset similarity threshold, determining that the current face matching result is successful; if at least one similarity value does not exceed a preset similarity threshold value, determining that the current face matching result is unsuccessful in matching;
the step of determining whether the current live broadcast room contains the real-name authentication anchor broadcast or not based on the current face matching result and the current face matching result comprises the following steps:
if the matching is successful, determining that the current live broadcast room comprises the real-name authentication anchor; otherwise, determining that the current live broadcast room does not contain the real-name authentication anchor.
4. The method of claim 1, wherein,
the step of acquiring the current face information in the current live broadcast room based on the video stream data of the current live broadcast room comprises the following steps:
acquiring a live video data stream;
performing frame extraction on the live video data stream according to a preset time interval to obtain a frame sequence;
and extracting face information from each frame image in the frame sequence, and sequentially using the face information as the current face information according to the frame extraction sequence.
5. The method of claim 1, wherein,
the step of obtaining the continuous time length of the current live broadcast room, which does not contain the real-name authentication anchor broadcast, and determining whether the illegal alternate broadcast situation occurs in the current live broadcast room based on the continuous time length comprises the following steps:
recording a time stamp of a video frame image corresponding to the current face information under the condition that the current live broadcast room does not contain a real-name authentication anchor broadcast;
acquiring continuous time length of a current live broadcast room, which does not contain a real-name authentication anchor, based on a timestamp of a current record and a timestamp of a historical record;
judging whether the continuous time length exceeds a preset time length threshold value or not; if so, determining that the illegal alternate broadcasting situation occurs in the current direct broadcasting room; and if not, determining that the illegal multicast condition does not occur in the current live broadcast room.
6. A detection apparatus of a live room anchor, comprising:
the system comprises a current face information acquisition module, a face information acquisition module and a face information acquisition module, wherein the current face information acquisition module is used for acquiring current face information in a current live broadcast room based on video stream data of the current live broadcast room;
the face matching result acquisition module is used for carrying out face matching on the current face information, the pre-stored face information of the real-name authentication anchor broadcast of the current live broadcast room and the pilot broadcast face information to obtain a current face matching result;
the real-name authentication anchor confirming module is used for determining whether the real-name authentication anchor is contained in the current live broadcast room or not based on the current face matching result;
and the illegal multicast generation judging module is used for obtaining the continuous time length of the current live broadcast room, which does not contain the real-name authentication anchor broadcast, and determining whether the illegal multicast generation condition occurs in the current live broadcast room or not based on the continuous time length.
7. The apparatus of claim 6, wherein,
and the face matching result acquisition module is used for carrying out face matching on the current face information, the face information of the real-name authentication anchor broadcast of the current live broadcast room, pilot broadcast face information and at least one piece of stored historical live broadcast face information to obtain a current face matching result.
8. The apparatus of claim 7, wherein,
the face matching result acquisition module is used for respectively calculating each similarity between the current face information and pre-stored real-name authentication face information, pilot broadcast face information and each historical live broadcast face information of the live broadcast room;
if all the similarity exceeds a preset similarity threshold, determining that the current face matching result is successful; if at least one similarity value does not exceed a preset similarity threshold value, determining that the current face matching result is unsuccessful in matching;
the real-name authentication anchor confirming module is used for determining that the current live broadcast room comprises the real-name authentication anchor if the matching is successful; otherwise, determining that the current live broadcast room does not contain the real-name authentication anchor.
9. The apparatus of claim 6, wherein,
the current face information acquisition module is used for acquiring a live video data stream;
performing frame extraction on the live video data stream according to a preset time interval to obtain a frame sequence;
and extracting face information from each frame image in the frame sequence, and sequentially using the face information as the current face information according to the frame extraction sequence.
10. The apparatus of claim 6, wherein,
the illegal substituted broadcasting judging module is used for recording a time stamp of a video frame image corresponding to the current face information under the condition that the current live broadcasting room does not contain a real-name authentication anchor broadcasting;
acquiring continuous time length of a current live broadcast room, which does not contain a real-name authentication anchor, based on a timestamp of a current record and a timestamp of a historical record;
judging whether the continuous time length exceeds a preset time length threshold value or not; if so, determining that the illegal alternate broadcasting situation occurs in the current direct broadcasting room; and if not, determining that the illegal multicast condition does not occur in the current live broadcast room.
11. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5.
12. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-5.
13. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111268834.7A CN113961898A (en) | 2021-10-29 | 2021-10-29 | Detection method, device and equipment for anchor in live broadcast room and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111268834.7A CN113961898A (en) | 2021-10-29 | 2021-10-29 | Detection method, device and equipment for anchor in live broadcast room and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113961898A true CN113961898A (en) | 2022-01-21 |
Family
ID=79468282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111268834.7A Pending CN113961898A (en) | 2021-10-29 | 2021-10-29 | Detection method, device and equipment for anchor in live broadcast room and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113961898A (en) |
-
2021
- 2021-10-29 CN CN202111268834.7A patent/CN113961898A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104539514B (en) | Information filtering method and device | |
CN113205037B (en) | Event detection method, event detection device, electronic equipment and readable storage medium | |
CN112770129A (en) | Live broadcast-based group chat establishing method, related device, equipment and medium | |
CN112784760A (en) | Human behavior recognition method, device, equipment and storage medium | |
CN114783061B (en) | Smoking behavior detection method, device, equipment and medium | |
CN115019390A (en) | Video data processing method and device and electronic equipment | |
CN113011345A (en) | Image quality detection method and device, electronic equipment and readable storage medium | |
CN113961898A (en) | Detection method, device and equipment for anchor in live broadcast room and storage medium | |
CN111783515A (en) | Behavior action recognition method and device | |
CN115329314A (en) | Face authentication method, device, equipment, system, storage medium and product | |
CN115270984A (en) | Anti-fraud model construction method and device, electronic equipment and storage medium | |
CN114884696A (en) | Fraud early warning method, device, equipment and storage medium | |
CN114550300A (en) | Video data analysis method and device, electronic equipment and computer storage medium | |
CN113852835A (en) | Live broadcast audio processing method and device, electronic equipment and storage medium | |
CN113011301A (en) | Living body identification method and device and electronic equipment | |
CN112995018A (en) | Message processing method, device, equipment and storage medium | |
CN112863548A (en) | Method for training audio detection model, audio detection method and device thereof | |
CN106339376B (en) | Method and device for identifying hot microblog | |
CN105701684B (en) | Data processing method and device | |
CN117119434B (en) | Personnel identification method, device, equipment and storage medium | |
CN115603947A (en) | Abnormal access detection method and device | |
CN114339642B (en) | Identification method, device, equipment and medium for accompanying user | |
CN114241374B (en) | Training method of live broadcast processing model, live broadcast processing method, device and equipment | |
CN117421199B (en) | Behavior determination method and system | |
CN114048058A (en) | Live event searching method and device, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |