CN118042175B - Live broadcast room data generation method and device, storage medium and electronic device - Google Patents
Live broadcast room data generation method and device, storage medium and electronic device Download PDFInfo
- Publication number
- CN118042175B CN118042175B CN202410145200.XA CN202410145200A CN118042175B CN 118042175 B CN118042175 B CN 118042175B CN 202410145200 A CN202410145200 A CN 202410145200A CN 118042175 B CN118042175 B CN 118042175B
- Authority
- CN
- China
- Prior art keywords
- user
- data
- live broadcast
- selection interface
- activation operation
- 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.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/21—Server components or server architectures
- H04N21/218—Source of audio or video content, e.g. local disk arrays
- H04N21/2187—Live feed
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4312—Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content
- H04N21/47202—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content for requesting content on demand, e.g. video on demand
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/482—End-user interface for program selection
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Databases & Information Systems (AREA)
- Collating Specific Patterns (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention relates to the technical field of live broadcast data processing, in particular to a live broadcast room data generation method. The method comprises the following steps: responding to the activation operation of the user, and generating a live broadcast selection interface; and responding to the touch operation of the user on the live broadcast selection interface, and generating live broadcast room interface data. The invention realizes the deep understanding of the user behavior, thereby improving the flexibility of live broadcast data statistics.
Description
Technical Field
The present invention relates to the field of live broadcast data processing technologies, and in particular, to a method and an apparatus for generating live broadcast data, a storage medium, and an electronic device.
Background
Live platforms are one of the most popular entertainment platforms at the time. The live broadcast platform is used for providing live broadcast content of the anchor to the audience, and generating live broadcast statistical data of the live broadcast platform according to interaction information of the audience and the anchor in a live broadcast room of the anchor.
The live broadcast server of the live broadcast platform generally obtains interaction information of a live broadcast room according to information generated by interaction between a viewer and a host in the live broadcast room of the host in a client side of the viewer. The interactive information includes information that the viewer joins the living room, information that the viewer transmits a barrage comment at the living room, and information that the viewer gives a gift to the anchor. And then, the live broadcast server generates live broadcast statistical data of the live broadcast platform according to the interaction information of all live broadcast rooms in the live broadcast platform. The live statistics are used to reflect the operational status of the live platform, such as the revenue of the live platform, the number of viewers of the live platform, and the active proportion of viewers of the live platform.
The live broadcast server only can generate live broadcast statistical data of the live broadcast platform according to the interaction information of all live broadcast rooms in the live broadcast platform, and statistics cannot be included for the operation condition of a user, so that the flexibility of generating the live broadcast statistical data is low.
Disclosure of Invention
The invention provides a method for generating live broadcast room data to solve the technical problem of low statistical flexibility of live broadcast data.
The application provides a method for generating data in a live broadcasting room, which comprises the following steps:
S1: responding to the activation operation of the user, and generating a live broadcast selection interface; the activation operation at least comprises any one of the user activation operation, the third party activation operation and the security activation operation, and the live broadcast selection interface at least comprises any one of a user live broadcast selection interface, a third party live broadcast selection interface and a security live broadcast selection interface; s2: and responding to the touch operation of the user on the live broadcast selection interface, and generating live broadcast room interface data.
Optionally, the generating the live selection interface in response to the activation operation of the user includes:
s11: responding to user activation operation of a user, and generating a user live broadcast selection interface;
s12: or responding to the third party activation operation of the user, and generating a third party live broadcast selection interface;
s13: or responding to the safe activation operation of the user, generating a safe live broadcast selection interface, wherein the authority level data corresponding to the user live broadcast selection interface is larger than the authority level data corresponding to the third-party live broadcast selection interface, and the authority level data corresponding to the third-party live broadcast selection interface is larger than the authority level data corresponding to the safe live broadcast selection interface.
According to the method and the device, a user live broadcast selection interface is generated through a deep learning model or a rule engine according to the activation operation of the user, and the interface reflects interests, preferences and historical viewing records of the user and provides personalized live broadcast recommendation. And for the third party activation operation, generating a live broadcast selection interface based on the data access of the third party platform. And considering the characteristics of the third party platform, integrating the third party live contents, providing diversified live broadcast selection, and simultaneously avoiding the problem of privacy disclosure caused by personalized recommendation. Aiming at the security activation operation, a security live broadcast selection interface is generated, so that live broadcast content is ensured to meet security standards, and the problems of data leakage or improper operation caused by too high authority of non-users or other modes are reduced. The method ensures that the user obtains more personalized and diversified watching experience through ingenious generation of the live broadcast selection interface, simultaneously emphasizes the safety and compliance, and improves the competitiveness of the platform and the satisfaction of the user.
Optionally, the activating operation in response to the user includes:
s101: responding to the activation operation of a user, and acquiring user gesture data and user fingerprint data;
S102: when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is pre-stored user fingerprint data, the activation operation is regarded as user activation operation, and a user live broadcast selection interface is generated;
S103: when the user gesture data is determined to be second pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data, or when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, the activation operation is regarded as a third party activation operation, and a third party live broadcast selection interface is generated;
s104: and when the user gesture data is determined to be third pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data or when the user gesture data is determined to be non-pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, generating a safe live broadcast selection interface.
By collecting the user gesture data and the user fingerprint data, the invention can realize the highly personalized identification of the user activation operation, and the collection of the user gesture and the fingerprint data can establish the unique biological characteristics and behavior habits of the user, thereby improving the accuracy of the activation operation. By adopting a mode of acquiring the gesture and fingerprint data of the user in real time, the user can quickly make judgment while performing activation operation, the instantaneity and instantaneity are realized, the response speed of the system is improved, and smoother operation experience is provided for the user. The activation operation is divided into user activation operation, third party activation operation and safe activation operation through different judgment conditions, multi-level activation operation classification is achieved, operation of a user and a third party is facilitated to be distinguished, different live broadcast selection interfaces are provided, and requirements of different users are met better. Through verifying the user gesture and the fingerprint data, whether the user gesture and the fingerprint data are prestored or not can be judged, safety is enhanced, and when the user gesture and the fingerprint data which are not prestored are identified, safety activation operation is adopted, so that the safety of the platform is ensured, and illegal access is prevented. The user does not need to manually switch the interface, the system actively generates a corresponding live broadcast selection interface according to the activation mode of the user, the operation burden of the user is reduced, and the convenience is improved. Through verification of user gestures and fingerprint data, the probability of misoperation is effectively reduced, and for judgment of a third party or non-prestored user data, activation errors caused by misidentification can be avoided, and the stability of the system is improved.
Optionally, before generating the user live broadcast selection interface, the method further comprises the following steps:
S111: starting a camera to acquire user scene data to obtain the user scene data;
s112: face detection is carried out on the user scene data to obtain face detection data;
S113: when the face detection data are determined to be single face detection data, face recognition is carried out on the user scene data by utilizing the face detection data, so that face recognition data are obtained;
S114: when the face recognition data are determined to be prestored user face data, generating a user live broadcast selection interface;
s115: and when the face recognition data is determined to be non-prestored user face data or the face detection data is determined to be non-single face detection data, executing the step S103 to regard the activation operation as a third party activation operation, and generating a third party live broadcast selection interface.
In the invention, the S112 carries out face detection on the user scene data, can accurately determine the face position and angle of the user, is beneficial to eliminating the interference of a non-face area and improves the accuracy of the subsequent face recognition. S113 and S114 ensure that a user live broadcast selection interface is generated only when a single face is detected and the identification of pre-stored user face data is completed, so that the stronger verification of the user identity is increased, and misoperation of multiple users or non-pre-stored users is prevented. By using pre-stored user face data to perform face recognition (S114), it is ensured that the identity of the user is verified, a user live broadcast selection interface is generated, unauthorized user access is prevented, and generation of the live broadcast selection interface is ensured to be performed only under legal user identities. In S115, when the non-single face or non-prestored user face data is detected, the activation operation is regarded as a third party activation operation, and a third party live broadcast selection interface is generated. By face detection and recognition, the method and the device compare with prestored user face data, double verification of user identities is achieved, calculation load of a system is reduced, meanwhile, further face recognition is conducted according to the face detection result, accuracy is improved, the generated live broadcast selection interface is guaranteed to be aimed at legal users, and safety of a platform is improved.
Optionally, the performing face detection on the user scene data to obtain face detection data includes:
s1121: carrying out parameter correction on a preset texture feature extractor by using user scene data to obtain a texture feature optimization extractor;
S1122: extracting texture features of the user scene data by using a texture feature optimization extractor to obtain texture feature data, and extracting color features of the user scene data to obtain texture feature data and color feature data;
s1123: performing attention calculation on the texture feature data and the color feature data to obtain texture self-attention feature data and color self-attention feature data;
s1124: pooling layer calculation is carried out on the texture self-attention characteristic data and the color self-attention characteristic data to obtain texture pooling layer data and color pooling layer data;
S1125: and processing the face detection data by using preset target detection head data to obtain face detection data.
In the invention, the S1121 carries out parameter correction on the preset texture feature extractor, so that the texture feature extractor can be better adapted to different scenes and illumination conditions, the accuracy and stability of texture features can be improved, and the reliability of face detection can be improved. And S1122, texture and color feature extraction is performed on the user scene data to obtain texture feature data and color feature data, so that the extraction of multi-feature dimensions is helpful for describing the user scene more comprehensively, and the robustness of face detection is improved. S1123, performing attention calculation on the texture feature data and the color feature data to obtain texture self-attention feature data and color self-attention feature data, and introducing a self-attention mechanism, so that the system focuses on important features, and the sensitivity and the accuracy of detection are improved. S1124 performs pooling layer calculation on the self-attention feature data to obtain texture pooling layer data and color pooling layer data, and the processing is helpful for reducing feature dimensions, extracting more important information, reducing calculation complexity, and maintaining sensitivity to key information. S1125, processing face detection data on texture pooling layer data and color pooling layer data by utilizing preset target detection head data, and combining a target detection technology in the mode, locating and recognizing faces more accurately, so that the face detection accuracy is improved.
Optionally, the parameter correction is performed by using a user scene texture parameter correction calculation formula, where the user scene texture parameter correction calculation formula specifically includes:
θ' is parameter data of the texture feature optimization extractor, θ is parameter data of the preset texture feature optimization extractor, α is parameter correction step control item, L is texture feature loss control item, f i is ith feature output item of the texture feature optimization extractor, i is feature output sequence item of the texture feature optimization extractor, and n is feature output number item of the texture feature optimization extractor.
The invention constructs a calculation formula for correcting the texture parameters of the user scene, which is used for parameter updating and is applied to parameter correction of the texture feature optimization extractor in S1121. Wherein alpha is a parameter correction step length control item, and the step length of each update is controlled, which is a constant, and the amplitude of the parameter update is determined. L is a texture penalty control term, which represents a penalty function in the optimization process, and the design of this penalty function should be such that the optimization process can be advanced towards the goal of improving texture. The calculation formula applies the idea of gradient descent, and parameters of the texture feature optimization extractor are adjusted continuously and iteratively, so that the loss of the texture feature is gradually reduced, and the extractor can be better adapted to different user scenes. Gradient terms in the formulaThe rate of change of the loss function with respect to the parameter, i.e. the gradient, is shown, determining the direction of parameter modification, so that the optimization algorithm can update the parameter in a direction to reduce the loss. The adjustment of the step control term α affects the size of each parameter update, and the appropriate step may cause the optimization algorithm to converge faster, but too large a step may cause oscillations or miss the optimal solution. The design of the loss function L is critical in the optimization process. It should be able to accurately describe the merits of the texture feature and guide the optimization algorithm towards improving the texture feature. On the basis of the concept of the gradient descent method, the optimization target of the texture features is combined, and parameters can be continuously adjusted in the iteration process, so that the texture feature optimization extractor is better adapted to a user scene, and the accuracy of face detection is improved.
The invention has the beneficial effects that: the selection interface of the living broadcast room is generated through touch operation, the user is integrated with the experience of active selection from the traditional living broadcast watching, the touch technology is adopted, and the intuitive interface design is combined, so that the selection and participation of the user to the living broadcast room are enhanced, interactive data can be better generated according to the interaction condition of the user and the living broadcast room, and the flexibility of the data statistics of the living broadcast room is improved.
Drawings
Other features, objects and advantages of the application will become more apparent upon reading of the detailed description of a non-limiting implementation, made with reference to the accompanying drawings in which:
FIG. 1 illustrates a flow diagram of a method of live room data generation of an embodiment;
FIG. 2 illustrates a flow diagram of a method of generating live room interface data of an embodiment;
FIG. 3 is a flow chart of an activation operation determination method of an embodiment;
FIG. 4 is a flow chart illustrating a user live selection interface secondary judgment method according to an embodiment;
fig. 5 shows a flowchart of a face detection method based on a live room according to an embodiment.
Detailed Description
The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.
Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.
It will be understood that, although the terms "first," "second," etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Referring to fig. 1 to 5, the present application provides a method for generating data in a live broadcast room, which includes the following steps:
s1: responding to the activation operation of the user, and generating a live broadcast selection interface; the activation operation at least comprises any one of the user activation operation, the third party activation operation and the security activation operation, and the live broadcast selection interface at least comprises any one of a user live broadcast selection interface, a third party live broadcast selection interface and a security live broadcast selection interface;
s2: and responding to the touch operation of the user on the live broadcast selection interface, and generating live broadcast room interface data.
According to the invention, through the live broadcasting room selection interface generated by the S1, live broadcasting room interface data can be generated in real time according to the touch operation of the user, so that highly personalized user experience is provided, and meanwhile, the operation process of the user on the live broadcasting room is better analyzed, so that the statistical flexibility of the live broadcasting data is improved.
Optionally, the generating the live selection interface in response to the activation operation of the user includes:
s11: responding to user activation operation of a user, and generating a user live broadcast selection interface;
in one embodiment of the invention, in response to a user activation operation by a user activating an application for the user, the action Navigation is triggered to navigate to the user live selection interface. And constructing a user live selection interface by utilizing REACT NATIVE components, wherein the user live selection interface comprises a live list, user information display and the like. The user's activation operation may include clicking on an application icon, designating a gesture, etc.
S12: or responding to the third party activation operation of the user, and generating a third party live broadcast selection interface;
In one embodiment of the invention, in response to a third party activation operation of a user, a third party live broadcast selection interface is generated to activate an application for a third party partner, and the action Navigation is triggered to navigate to the third party live broadcast selection interface. A third party live selection interface is constructed using REACT NATIVE components, including partner live lists, recommended content, and the like. The third party activation operation is realized by means of API call, partner SDK and the like.
S13: or responding to the safe activation operation of the user, generating a safe live broadcast selection interface, wherein the live broadcast selection interface comprises a user live broadcast selection interface, a third-party live broadcast selection interface and a safe live broadcast selection interface, the authority level data corresponding to the user live broadcast selection interface is larger than the authority level data corresponding to the third-party live broadcast selection interface, and the authority level data corresponding to the third-party live broadcast selection interface is larger than the authority level data corresponding to the safe live broadcast selection interface;
In one embodiment of the invention, the security activation operation includes biometric authentication, password verification, and the like. And triggering the action Navigation to the safe live broadcast selection interface according to the safety authentication result. A secure live selection interface is constructed using REACT NATIVE components, including restricted content, secure filtering, and the like.
According to the method and the device, a user live broadcast selection interface is generated through a deep learning model or a rule engine according to the activation operation of the user, and the interface reflects interests, preferences and historical viewing records of the user and provides personalized live broadcast recommendation. And for the third party activation operation, generating a live broadcast selection interface based on the data access of the third party platform. And considering the characteristics of the third party platform, integrating the third party live contents, providing diversified live broadcast selection, and simultaneously avoiding the problem of privacy disclosure caused by personalized recommendation. Aiming at the security activation operation, a security live broadcast selection interface is generated, so that live broadcast content is ensured to meet security standards, and the problems of data leakage or improper operation caused by too high authority of non-users or other modes are reduced. According to the live broadcasting room data generation method, through ingenious generation of the live broadcasting selection interface, a user obtains more personalized and diversified watching experience, meanwhile, safety and compliance are emphasized, and the competitiveness of a platform and user satisfaction are improved.
Optionally, the activating operation includes a user activating operation, a third party activating operation and a security activating operation, and the judging mode of the activating operation specifically includes:
s101: responding to the activation operation of a user, and acquiring user gesture data and user fingerprint data;
In one embodiment of the invention, user gesture data is acquired in response to an activation operation by a user and user fingerprint data is captured by utilizing REACT NATIVE Gesture Handler libraries, such as swipes, clicks, etc. And calling a biological recognition library to acquire fingerprint data or other biological characteristic data of the user, so as to ensure the uniqueness of the identity.
In one embodiment of the invention, user gesture data refers to data generated by hand movements or gestures of a user on a device. Such data is typically captured by sensors (e.g., cameras, gyroscopes, accelerometers, etc.) for identifying and recording user-specific hand movements, finger movements, or gesture patterns. The gesture data includes information of: gesture type: the type of gesture is recognized, such as clicking, sliding, pinching, rotating, etc. Spatial location of gesture: the specific location and direction in which a gesture occurs is recorded, typically in a three-dimensional coordinate system. Time information of gesture: information related to the time point, duration and the like of the gesture is recorded. Motion trajectories of fingers or hands: track recordings are made of the movements of the finger or hand in order to analyze the fluency and speed of the gesture. Angle and gesture of gesture: and recording rotation and inclination information of the gesture and the like so as to acquire richer gesture characteristics.
S102: when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is pre-stored user fingerprint data, the activation operation is regarded as user activation operation, and a user live broadcast selection interface is generated;
In one embodiment of the invention, the user gesture data is determined to be first pre-stored user gesture data, the user fingerprint data is pre-stored user fingerprint data, hash encryption is performed on the user gesture data and the fingerprint data, the encrypted data is compared with the pre-stored user gesture data and the pre-stored fingerprint data, and if the matching is successful, the user activation operation is determined.
S103: when the user gesture data is determined to be second pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data, or when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, the activation operation is regarded as a third party activation operation, and a third party live broadcast selection interface is generated;
in one embodiment of the present invention, the user gesture data is second pre-stored user gesture data and the user fingerprint data is pre-stored user fingerprint data, the user gesture data and the fingerprint data are hashed and encrypted, the encrypted data and the second group of pre-stored user gesture data and the fingerprint data are compared, and if the matching is successful, the third party activation operation is determined.
S104: and when the user gesture data is determined to be third pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data or when the user gesture data is determined to be non-pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, generating a safe live broadcast selection interface.
In one embodiment of the invention, the user gesture data and the fingerprint data are hashed and encrypted, and the encrypted data are compared with a third set of pre-stored user gesture data and fingerprint data. And if the matching is successful, determining to be a safe activation operation, or determining to be a safe activation operation if the user gesture data and the fingerprint data are matched to be non-prestored user gesture data and fingerprint data.
By collecting the user gesture data and the user fingerprint data, the invention can realize the highly personalized identification of the user activation operation, and the collection of the user gesture and the fingerprint data can establish the unique biological characteristics and behavior habits of the user, thereby improving the accuracy of the activation operation. By adopting a mode of acquiring the gesture and fingerprint data of the user in real time, the user can quickly make judgment while performing activation operation, the instantaneity and instantaneity are realized, the response speed of the system is improved, and smoother operation experience is provided for the user. The activation operation is divided into user activation operation, third party activation operation and safe activation operation through different judgment conditions, multi-level activation operation classification is achieved, operation of a user and a third party is facilitated to be distinguished, different live broadcast selection interfaces are provided, and requirements of different users are met better. Through verifying the user gesture and the fingerprint data, whether the user gesture and the fingerprint data are prestored or not can be judged, safety is enhanced, and when the user gesture and the fingerprint data which are not prestored are identified, safety activation operation is adopted, so that the safety of the platform is ensured, and illegal access is prevented. The user does not need to manually switch the interface, the system actively generates a corresponding live broadcast selection interface according to the activation mode of the user, the operation burden of the user is reduced, and the convenience is improved. Through verification of user gestures and fingerprint data, the probability of misoperation is effectively reduced, and for judgment of a third party or non-prestored user data, activation errors caused by misidentification can be avoided, and the stability of the system is improved.
Optionally, before generating the user live broadcast selection interface, the method further comprises the following steps:
S111: starting a camera to acquire user scene data to obtain the user scene data;
in one embodiment of the invention, the device camera is started for real-time video acquisition by using an application program or system authority. And processing the collected video data, and extracting user scene data including user positions, environment backgrounds and the like.
In one embodiment of the invention, the user triggers a live function in the application, and the system requests and acquires camera rights. And opening the camera to acquire real-time video, and capturing the current scene of the user.
S112: face detection is carried out on the user scene data to obtain face detection data;
In one embodiment of the invention, a face detection algorithm is applied to the acquired user scene data to detect the face position in the image. And outputting face detection data comprising information such as coordinates, size and the like of the face.
In one embodiment of the invention, a face detection algorithm is applied to the acquired video frames to locate the faces in the image. And outputting face detection data comprising information such as coordinates, size and the like of the face.
S113: when the face detection data are determined to be single face detection data, face recognition is carried out on the user scene data by utilizing the face detection data, so that face recognition data are obtained;
In detail, the face detection data is checked, and if the number of detected faces is 1, it is considered as single face detection data.
S114: when the face recognition data are determined to be prestored user face data, generating a user live broadcast selection interface;
In one embodiment of the invention, a face recognition system is constructed using pre-stored user face data. And matching the detected face data with pre-stored face data of the user, and confirming whether the user is a known user or not.
S115: and when the face recognition data is determined to be non-prestored user face data or the face detection data is determined to be non-single face detection data, executing the step S103 to regard the activation operation as a third party activation operation, and generating a third party live broadcast selection interface.
In one embodiment of the present invention, if the face recognition data does not match pre-stored user face data or a plurality of faces are detected, error processing is performed. Triggering a third party activation operation according to specific conditions, and executing corresponding logic.
In the invention, the S112 carries out face detection on the user scene data, can accurately determine the face position and angle of the user, is beneficial to eliminating the interference of a non-face area and improves the accuracy of the subsequent face recognition. S113 and S114 ensure that a user live broadcast selection interface is generated only when a single face is detected and the identification of pre-stored user face data is completed, so that the stronger verification of the user identity is increased, and misoperation of multiple users or non-pre-stored users is prevented. By using pre-stored user face data to perform face recognition (S114), it is ensured that the identity of the user is verified, a user live broadcast selection interface is generated, unauthorized user access is prevented, and generation of the live broadcast selection interface is ensured to be performed only under legal user identities. In S115, when the non-single face or non-prestored user face data is detected, the activation operation is regarded as a third party activation operation, and a third party live broadcast selection interface is generated. By face detection and recognition, the method and the device compare with prestored user face data, double verification of user identities is achieved, calculation load of a system is reduced, meanwhile, further face recognition is conducted according to the face detection result, accuracy is improved, the generated live broadcast selection interface is guaranteed to be aimed at legal users, and safety of a platform is improved.
Optionally, the performing face detection on the user scene data to obtain face detection data includes:
s1121: carrying out parameter correction on a preset texture feature extractor by using user scene data to obtain a texture feature optimization extractor;
In one embodiment of the present invention, the user scene data is processed by using a preset texture feature extractor to obtain the original texture feature data. And designing a parameter correction calculation formula, wherein the parameter correction calculation formula can be optimized and adjusted by adopting a gradient descent or back propagation algorithm, and the formula can be used for adjusting parameters of the texture feature extractor according to the characteristics of the user scene data.
S1122: extracting texture features of the user scene data by using a texture feature optimization extractor to obtain texture feature data, and extracting color features of the user scene data to obtain texture feature data and color feature data;
In one embodiment of the invention, the optimized texture feature extractor is utilized to carry out convolution operation on the user scene data to extract texture feature data. And simultaneously extracting color characteristics to obtain color characteristic data.
S1123: performing attention calculation on the texture feature data and the color feature data to obtain texture self-attention feature data and color self-attention feature data;
In one embodiment of the invention, a self-attention mechanism is applied to the texture feature data and the color feature data, and the texture self-attention feature data and the color self-attention feature data are calculated. If the texture self-attention feature data is mapped to the weight matrix linearly, attention score calculation is carried out on the weight matrix, attention weight calculation is carried out on the obtained attention score data, weighted summation is carried out on the attention weight data, and the texture self-attention feature data and the color self-attention feature data are obtained.
S1124: pooling layer calculation is carried out on the texture self-attention characteristic data and the color self-attention characteristic data to obtain texture pooling layer data and color pooling layer data;
In one embodiment of the invention, the texture self-attention feature data and the color self-attention feature data are pooled using maximum pooling or average pooling.
S1125: and processing the face detection data by using preset target detection head data to obtain face detection data.
In one embodiment of the present invention, face detection is performed on texture pooling layer data and color pooling layer data using preset target detection head data. And outputting face detection data comprising information such as the position, confidence and the like of the face.
In the invention, the S1121 carries out parameter correction on the preset texture feature extractor, so that the texture feature extractor can be better adapted to different scenes and illumination conditions, the accuracy and stability of texture features can be improved, and the reliability of face detection can be improved. And S1122, texture and color feature extraction is performed on the user scene data to obtain texture feature data and color feature data, so that the extraction of multi-feature dimensions is helpful for describing the user scene more comprehensively, and the robustness of face detection is improved. S1123, performing attention calculation on the texture feature data and the color feature data to obtain texture self-attention feature data and color self-attention feature data, and introducing a self-attention mechanism, so that the system focuses on important features, and the sensitivity and the accuracy of detection are improved. S1124 performs pooling layer calculation on the self-attention feature data to obtain texture pooling layer data and color pooling layer data, and the processing is helpful for reducing feature dimensions, extracting more important information, reducing calculation complexity, and maintaining sensitivity to key information. S1125, processing face detection data on texture pooling layer data and color pooling layer data by utilizing preset target detection head data, and combining a target detection technology in the mode, locating and recognizing faces more accurately, so that the face detection accuracy is improved.
Optionally, the parameter correction is performed by using a user scene texture parameter correction calculation formula, where the user scene texture parameter correction calculation formula specifically includes:
θ' is parameter data of the texture feature optimization extractor, θ is parameter data of the preset texture feature optimization extractor, α is parameter correction step control item, L is texture feature loss control item, f i is ith feature output item of the texture feature optimization extractor, i is feature output sequence item of the texture feature optimization extractor, and n is feature output number item of the texture feature optimization extractor.
In one embodiment of the invention, the initial parameter values are: θ=2.0, learning rate α=0.01, and parameter correction is performed: θ' =2.02, and multiple iterations make it close to 3.
The invention constructs a calculation formula for correcting the texture parameters of the user scene, which is used for parameter updating and is applied to parameter correction of the texture feature optimization extractor in S1121. Wherein alpha is a parameter correction step length control item, and the step length of each update is controlled, which is a constant, and the amplitude of the parameter update is determined. L is a texture penalty control term, which represents a penalty function in the optimization process, and the design of this penalty function should be such that the optimization process can be advanced towards the goal of improving texture. The calculation formula applies the idea of gradient descent, and parameters of the texture feature optimization extractor are adjusted continuously and iteratively, so that the loss of the texture feature is gradually reduced, and the extractor can be better adapted to different user scenes. Gradient terms in the formulaThe rate of change of the loss function with respect to the parameter, i.e. the gradient, is shown, determining the direction of parameter modification, so that the optimization algorithm can update the parameter in a direction to reduce the loss. The adjustment of the step control term α affects the size of each parameter update, and the appropriate step may cause the optimization algorithm to converge faster, but too large a step may cause oscillations or miss the optimal solution. The design of the loss function L is critical in the optimization process. It should be able to accurately describe the merits of the texture feature and guide the optimization algorithm towards improving the texture feature. On the basis of the concept of the gradient descent method, the optimization target of the texture features is combined, and parameters can be continuously adjusted in the iteration process, so that the texture feature optimization extractor is better adapted to a user scene, and the accuracy of face detection is improved.
The direct broadcasting room selection interface is generated through touch operation, the user is integrated with the experience of active selection from the traditional direct broadcasting watching, and the selection and participation feeling of the user to the direct broadcasting room is enhanced by adopting the touch technology and combining the visual interface design. The invention provides more intelligent and personalized live room experience, so that the user can participate in the live content more actively, and the user retention rate and the platform liveness are improved.
The embodiment also provides a live broadcast room data generating device, which is used for realizing the above embodiment and the preferred implementation manner, and the description is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
Specifically, the live broadcasting room data generating device comprises:
The selection interface generation module is used for responding to the activation operation of the user and generating a live broadcast selection interface; the activation operation at least comprises any one of the user activation operation, the third party activation operation and the security activation operation, and the live broadcast selection interface at least comprises any one of a user live broadcast selection interface, a third party live broadcast selection interface and a security live broadcast selection interface;
And the interface data generation module is used for responding to the touch operation of the user on the interface selected by the live broadcasting room and generating live broadcasting room interface data.
In an alternative embodiment, the generating the live selection interface in response to the activation operation of the user includes: s11: responding to user activation operation of a user, and generating a user live broadcast selection interface;
s12: or responding to the third party activation operation of the user, and generating a third party live broadcast selection interface;
S13: or responding to the safe activation operation of the user, and generating a safe live broadcast selection interface, wherein the authority level data corresponding to the user live broadcast selection interface is larger than the authority level data corresponding to the third-party live broadcast selection interface, and the authority level data corresponding to the third-party live broadcast selection interface is larger than the authority level data corresponding to the safe live broadcast selection interface.
Embodiments of the present invention also provide a computer readable storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.
In one exemplary embodiment, the computer readable storage medium may include, but is not limited to: a usb disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory RAM), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing a computer program.
An embodiment of the invention also provides an electronic device comprising a memory having stored therein a computer program and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.
In an exemplary embodiment, the electronic apparatus may further include a transmission device connected to the processor, and an input/output device connected to the processor.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (4)
1. A method for generating data in a live broadcast room, the method comprising:
s1: responding to the activation operation of the user, and generating a live broadcast selection interface; the live broadcast selection interface at least comprises any one of a user live broadcast selection interface, a third-party live broadcast selection interface and a safe live broadcast selection interface;
S2: responding to the touch operation of the user on the live broadcast selection interface, and generating live broadcast room interface data;
The generating a live selection interface in response to an activation operation of a user comprises:
s11: responding to user activation operation of a user, and generating a user live broadcast selection interface;
s12: or responding to the third party activation operation of the user, and generating a third party live broadcast selection interface;
S13: or responding to the safe activation operation of the user, generating a safe live broadcast selection interface, wherein the authority level data corresponding to the user live broadcast selection interface is larger than the authority level data corresponding to the third-party live broadcast selection interface, and the authority level data corresponding to the third-party live broadcast selection interface is larger than the authority level data corresponding to the safe live broadcast selection interface;
The activation operation in response to the user comprises the following steps:
s101: responding to the activation operation of a user, and acquiring user gesture data and user fingerprint data;
S102: when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is pre-stored user fingerprint data, the activation operation is regarded as user activation operation, and a user live broadcast selection interface is generated;
S103: when the user gesture data is determined to be second pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data, or when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, the activation operation is regarded as a third party activation operation, and a third party live broadcast selection interface is generated;
s104: and when the user gesture data is determined to be third pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data or when the user gesture data is determined to be non-pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, the activation operation is regarded as a safe activation operation, and a safe live broadcast selection interface is generated.
2. A live room data generation apparatus, comprising:
The selection interface generation module is used for responding to the activation operation of the user and generating a live broadcast selection interface; the live broadcast selection interface at least comprises any one of a user live broadcast selection interface, a third-party live broadcast selection interface and a safe live broadcast selection interface;
The interface data generation module is used for responding to the touch operation of the user on the live broadcast selection interface and generating live broadcast room interface data;
The generating a live selection interface in response to an activation operation of a user comprises:
s11: responding to user activation operation of a user, and generating a user live broadcast selection interface;
s12: or responding to the third party activation operation of the user, and generating a third party live broadcast selection interface;
S13: or responding to the safe activation operation of the user, generating a safe live broadcast selection interface, wherein the authority level data corresponding to the user live broadcast selection interface is larger than the authority level data corresponding to the third-party live broadcast selection interface, and the authority level data corresponding to the third-party live broadcast selection interface is larger than the authority level data corresponding to the safe live broadcast selection interface;
The activation operation in response to the user comprises the following steps:
s101: responding to the activation operation of a user, and acquiring user gesture data and user fingerprint data;
S102: when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is pre-stored user fingerprint data, the activation operation is regarded as user activation operation, and a user live broadcast selection interface is generated;
S103: when the user gesture data is determined to be second pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data, or when the user gesture data is determined to be first pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, the activation operation is regarded as a third party activation operation, and a third party live broadcast selection interface is generated;
s104: and when the user gesture data is determined to be third pre-stored user gesture data and the user fingerprint data is determined to be pre-stored user fingerprint data or when the user gesture data is determined to be non-pre-stored user gesture data and the user fingerprint data is determined to be non-pre-stored user fingerprint data, the activation operation is regarded as a safe activation operation, and a safe live broadcast selection interface is generated.
3. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program, wherein the computer program is arranged to execute the method of claim 1 when run.
4. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410145200.XA CN118042175B (en) | 2024-02-01 | 2024-02-01 | Live broadcast room data generation method and device, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410145200.XA CN118042175B (en) | 2024-02-01 | 2024-02-01 | Live broadcast room data generation method and device, storage medium and electronic device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118042175A CN118042175A (en) | 2024-05-14 |
| CN118042175B true CN118042175B (en) | 2024-09-27 |
Family
ID=90985173
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410145200.XA Active CN118042175B (en) | 2024-02-01 | 2024-02-01 | Live broadcast room data generation method and device, storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118042175B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599029A (en) * | 2019-09-10 | 2019-12-20 | 深圳市兴海物联科技有限公司 | Property management property end system |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI584146B (en) * | 2014-08-29 | 2017-05-21 | 鴻海精密工業股份有限公司 | Login system and method based on face recognition |
| CN106406931A (en) * | 2016-08-25 | 2017-02-15 | 广州华多网络科技有限公司 | Studio rapid starting method and device in application program, and terminal equipment |
| CN106453245B (en) * | 2016-08-30 | 2019-11-15 | 北京小米移动软件有限公司 | Verify the method and device of identity |
| CN107920282A (en) * | 2017-11-22 | 2018-04-17 | 广州虎牙信息科技有限公司 | Direct broadcasting room main broadcaster bean vermicelli ranks method, apparatus and corresponding terminal |
| CN110662091B (en) * | 2018-06-29 | 2022-03-25 | 武汉斗鱼网络科技有限公司 | Third-party live video access method, storage medium, electronic device and system |
| CN109688480A (en) * | 2019-01-14 | 2019-04-26 | 广州虎牙信息科技有限公司 | A kind of live broadcasting method, terminal device and storage medium |
-
2024
- 2024-02-01 CN CN202410145200.XA patent/CN118042175B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110599029A (en) * | 2019-09-10 | 2019-12-20 | 深圳市兴海物联科技有限公司 | Property management property end system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118042175A (en) | 2024-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11330012B2 (en) | System, method, and device of authenticating a user based on selfie image or selfie video | |
| US10242364B2 (en) | Image analysis for user authentication | |
| CN107018121B (en) | User identity authentication method and device | |
| US10698998B1 (en) | Systems and methods for biometric authentication with liveness detection | |
| CN111274928B (en) | Living body detection method and device, electronic equipment and storage medium | |
| US10635795B2 (en) | Dynamic graphic eye-movement authentication system and method using face authentication or hand authentication | |
| US9049983B1 (en) | Ear recognition as device input | |
| US20170169304A1 (en) | Method and apparatus for liveness detection | |
| CN113348431B (en) | Multi-factor authentication for virtual reality | |
| KR20070027629A (en) | Personal gesture sign | |
| KR101884291B1 (en) | Display apparatus and control method thereof | |
| KR102392437B1 (en) | Reflection-based control activation | |
| Xiao et al. | Secure behavioral biometric authentication with leap motion | |
| Goswami et al. | Face recognition captcha | |
| CN108594995A (en) | Electronic equipment operation method based on gesture recognition and electronic equipment | |
| CN118042175B (en) | Live broadcast room data generation method and device, storage medium and electronic device | |
| WO2019095097A1 (en) | Identity authentication method for electronic terminal, and electronic terminal | |
| CN114547581A (en) | Method and device for providing verification code system | |
| Powell et al. | A multibiometrics-based CAPTCHA for improved online security | |
| Nguyen et al. | Penetration vision through virtual reality headsets: identifying 360-degree videos from head movements | |
| CN114202677A (en) | Method and system for authenticating an occupant in a vehicle interior | |
| Park et al. | A hand posture recognition system utilizing frequency difference of infrared light | |
| US20250037509A1 (en) | System and method for determining liveness using face rotation | |
| WO2021260314A1 (en) | Device and method for authenticating a user of a first electronic device connected to a second electronic device | |
| Luo | Surveying face liveness detection in the deepfake era |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |