CN111741364A - Content accurate pushing method and system based on face recognition - Google Patents

Content accurate pushing method and system based on face recognition Download PDF

Info

Publication number
CN111741364A
CN111741364A CN202010569981.7A CN202010569981A CN111741364A CN 111741364 A CN111741364 A CN 111741364A CN 202010569981 A CN202010569981 A CN 202010569981A CN 111741364 A CN111741364 A CN 111741364A
Authority
CN
China
Prior art keywords
data
user
library
media asset
picture
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
Application number
CN202010569981.7A
Other languages
Chinese (zh)
Inventor
宋博然
段立新
何宜兵
张神力
蔡忠鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Tianhai Chenguang Technology Co ltd
Original Assignee
Shenzhen Tianhai Chenguang Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Tianhai Chenguang Technology Co ltd filed Critical Shenzhen Tianhai Chenguang Technology Co ltd
Priority to CN202010569981.7A priority Critical patent/CN111741364A/en
Publication of CN111741364A publication Critical patent/CN111741364A/en
Pending legal-status Critical Current

Links

Images

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]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/757—Matching configurations of points or features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16—Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168—Feature extraction; Face representation
    • 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
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
    • H04N21/25866—Management of end-user data
    • H04N21/25891—Management of end-user data being end-user preferences
    • 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
    • H04N21/41—Structure of client; Structure of client peripherals
    • H04N21/422—Input-only peripherals, i.e. input devices connected to specially adapted client devices, e.g. global positioning system [GPS]
    • H04N21/4223—Cameras
    • 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
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/4508—Management of client data or end-user data
    • H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences
    • 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
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4661—Deriving a combined profile for a plurality of end-users of the same client, e.g. for family members within a home
    • 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
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Molecular Biology (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Human Computer Interaction (AREA)
  • Computer Graphics (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a content accurate pushing method and system based on face recognition; wherein the method comprises the following steps: starting up and collecting video or picture data; intelligently analyzing the collected video or picture data and obtaining intelligent analysis result structured data; matching the intelligent analysis result with an original picture library to obtain the current user role; and calling the corresponding media asset library according to the current user role, and displaying the media asset library to the user. The invention further provides a content accurate pushing system based on face recognition. The content accurate pushing method and system based on face recognition can apply a machine vision face recognition technology, realize accurate pushing of the content and greatly improve user experience.

Description

Content accurate pushing method and system based on face recognition
Technical Field
The invention relates to the field of machine vision application, in particular to a content accurate pushing method and system based on face recognition.
Background
In the field of television application, IPTV or OTT products have already come into thousands of households, and the habit of watching television by users is that operators push corresponding EPG pages, and users select to watch, or users search for specific programs according to their own interests. Under the implementation of the prior art, one or more EPG pages are pushed to a family, and the program content cannot be pushed accurately based on the viewing habits of users.
Disclosure of Invention
In order to solve the problems that EPG (electronic program guide) presented contents are uniform when a user watches TV and user experience is poor in the prior art, the invention provides the content accurate pushing method and system based on face recognition, which can apply a machine vision face recognition technology to realize accurate pushing of the contents and greatly improve the user experience.
The technical scheme provided by the invention is as follows:
a content accurate pushing method based on face recognition is disclosed, wherein the method comprises the following steps:
starting up and collecting video or picture data.
And intelligently analyzing the acquired video or picture data and obtaining intelligent analysis result structured data.
And matching the intelligent analysis result with an original picture library to obtain the current user role.
And calling the corresponding media asset library according to the current user role, and displaying the media asset library to the user.
The content accurate pushing method based on face recognition, wherein the starting-up and the acquisition of video or picture data specifically comprise:
the startup refers to startup of equipment with a media playing function, such as a set top box or a television.
The set top box or the television is provided with a video or picture data acquisition module and is used for acquiring video or picture data.
The video or picture acquisition module includes but is not limited to a camera or a camera-like device with a video or picture acquisition function.
The acquisition of the video or the picture refers to that the video or picture acquisition module acquires video or picture data when the set top box or the television and other devices are started.
The content accurate pushing method based on face recognition comprises the following steps of intelligently analyzing acquired video or picture data to obtain structural data of an intelligent analysis result, and specifically comprises the following steps:
the acquired data is video data, and frame extraction processing needs to be carried out on the video data to obtain picture data.
The intelligent analysis refers to the intelligent analysis of the picture data obtained by frame extraction or the picture data obtained by collection.
And the intelligent analysis is to perform target detection and identification on the picture data and identify the face image appearing in the picture.
And the step of recognizing the face image also comprises the step of extracting the structural data of the data characteristics of the face image to obtain the structural data of the intelligent analysis result.
The machine vision recognition algorithm of the target detection and recognition model is based on a deep convolutional neural network.
The content accurate pushing method based on face recognition is characterized in that the method matches the enabled analysis result with an original picture library to obtain the current user role, and specifically comprises the following steps:
the intelligent analysis result refers to the structural data description of the human face features.
The original photo gallery is a photo gallery of family members which are entered by a user in advance.
The matching with the original picture library is actually matching with the structural data description of the face features in the original picture library.
And the matching outputs matching result information according to the similarity.
The matching result is one or more of the family members, or not the family members.
The content accurate pushing method based on face recognition, wherein the corresponding media asset library is called according to the current user role and is displayed to a user, and the method specifically comprises the following steps:
and calling a media asset library corresponding to the current user role as one of the family members, and displaying the EPG content to the user.
The current user role is several of the family members, and before the media asset library is called, the method also comprises the steps of comparing the priorities preset in the system by the several family members, selecting the media asset library of the family member with the high priority, and calling and displaying the EPG content to the user.
And calling the content of the media assets in the visitor mode and displaying EPG information to the user when the current user role is not a family member.
The content accurate pushing method based on face recognition, wherein the advanced entry of the original picture library specifically comprises the following steps:
the set top box or the television and other devices collect image information of the home user.
And extracting the characteristics of the user image information.
Matching the image information of different family users with the extracted characteristic information, and storing to form an original picture library.
After the family member picture library information is input, priority information of different family members is required to be set.
The content accurate pushing method based on face recognition specifically comprises the following steps:
when the user watches the television program content, the set top box or the television and other equipment can automatically acquire the program content information watched by the user.
The media server classifies the program content information watched by the user and acquires the label.
And the media server forms the acquired labels into a label library.
And the media server summarizes the media asset contents associated with the tag library to form a media asset library associated with the user.
The invention also discloses a content accurate pushing system based on face recognition, wherein the system specifically comprises:
the set-top box or the television and other devices with the media playing function are used for collecting data and playing media contents.
And the media server is used for collecting user data and providing accurate pushing of media asset content.
The content accurate pushing system based on face recognition, wherein the set-top box or television and other equipment specifically comprises:
and the data acquisition unit is used for acquiring video or image data of the human face.
And the face recognition unit is used for carrying out face recognition by the user and comparing the face recognition with the original picture library.
And the original picture library is used for storing original picture information of the home user.
And the media asset acquisition unit is used for acquiring the media asset data.
And the media asset presenting unit is used for presenting the media asset data.
The content accurate pushing system based on face recognition, wherein the media server specifically comprises:
and the user information acquisition unit is used for acquiring the viewing habit data of the user.
And the media asset classification unit is used for forming media asset classifications corresponding to the users according to the viewing habits of the users.
And the media asset library is used for storing the media asset audio and video files and the metadata files.
And the media asset pushing unit is used for pushing media asset data to the user.
The content accurate pushing method and system based on face recognition can apply a machine vision face recognition technology, realize accurate pushing of the content and greatly improve user experience.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments are briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a content accurate pushing method based on face recognition according to a preferred embodiment of the present invention.
Fig. 2 is a flowchart of an original photo library entry according to an embodiment of the present invention.
Fig. 3 is a flowchart of an embodiment of creating a media asset library corresponding to a user role according to the method for accurately pushing content based on face recognition.
Fig. 4 is a system architecture diagram of a content accurate pushing system based on face recognition according to the present invention.
Fig. 5 is a functional structure block diagram of a media playing device in a system architecture of a content accurate pushing system based on face recognition according to the present invention.
Fig. 6 is a functional structure block diagram of a media server in the system architecture of the content accurate pushing system based on face recognition according to the present invention.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and clearer, the present invention is described in further detail below. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a flow chart of an optimal embodiment of a content accurate pushing method based on face recognition, which is shown in figure 1. The method comprises the following specific steps:
step S101: starting up and collecting video or picture data.
The startup refers to startup of equipment with a media playing function, such as a set top box or a television.
The set top box or the television is provided with a video or picture data acquisition module and is used for acquiring video or picture data.
The video or picture acquisition module includes but is not limited to a camera or a camera-like device with a video or picture acquisition function.
The acquisition of the video or the picture refers to that the video or picture acquisition module acquires video or picture data when the set top box or the television and other devices are started.
The acquisition refers to shooting real-time video or picture data.
The collected video data may be h.264 or h.265 according to the specification and performance of the video collection module.
The captured picture data includes, but is not limited to: JPEG, JPEG2000, BMP.
In order to protect the family privacy of the user, the video or image acquisition module is started when the set top box or the television is started; when the system is turned off, the privacy of the user is protected by adopting a physical shade mode.
Step S102: and intelligently analyzing the acquired video or picture data and obtaining intelligent analysis result structured data.
And if the acquired data is video data, performing frame extraction processing on the video data to obtain picture data.
The frame extraction processing specifically refers to a process of decoding and then encoding the video data.
The format of the picture data obtained by frame extraction includes but is not limited to: JPEG, JPEG2000, BMP.
The intelligent analysis refers to the intelligent analysis of the picture data obtained by frame extraction or the picture data obtained by collection.
And the intelligent analysis is to perform target detection and identification on the picture data and identify the face image appearing in the picture.
And the step of recognizing the face image also comprises the step of extracting the structural data of the data characteristics of the face image to obtain the structural data of the intelligent analysis result.
The structured data of the face image refers to the extraction of feature points based on the face image.
The facial image feature points include but are not limited to: vertex, glabellar, occipital point, temporal lobe cover, midpoint of hair track, zygomatic point, inferior mandibular angle point, sublabial, submental point, mandibular point, external condyloid process point, external angle of the eye, center of pupil, lowest orbital point, upper portion of the eyelid, lower portion of the eyelid, highest orbital point, above-eyebrow point, nasion point, inferior nasion point, alar end, nasal protuberant point, inferior nasion point, superior alar end, alar curve, maxillofrontal point, superior labial point, inferior labial point, mouth horn, oral point, superior auricular point, inferior auricular point, anterior auricular point, posterior auricular point, superior auricular point, inferior auricular point, superior auricular point, middle auricular midpoint, superior auricular median superior auricular point, and tragus point;
the machine vision recognition algorithm of the target detection and recognition model is based on a deep convolutional neural network.
Step S103: and matching the intelligent analysis result with an original picture library to obtain the current user role.
The intelligent analysis result refers to the structural data description of the human face features.
The structured data refers to the extraction of feature points based on a face image.
The feature points of the face image are already described in detail in step S102.
The original photo gallery is a photo gallery of family members which are entered by a user in advance.
The original picture library comprises the related information of the original pictures of the family members.
The family member original picture related information includes but is not limited to: family member name, role, gender, age, face original picture, face picture structural description information, priority level and associated media asset information.
The matching with the original picture library is actually matching with the structural data description of the face features in the original picture library.
The structural description information is actually extracted based on the feature points of the face image.
And the matching outputs matching result information according to the similarity.
The similarity is preset with a threshold, and when the similarity of the matching results reaches the threshold, the same person is judged; if the threshold is not reached, it is determined as a different person.
The matching result is one or more of the family members, or not the family members.
Step S104: and calling the corresponding media asset library according to the current user role, and displaying the media asset library to the user.
And calling a media asset library corresponding to the current user role as one of the family members, and displaying the EPG content to the user.
The media asset library corresponds to family members.
The media resource library is an intelligently produced media resource library and is a set of content resources interested by the current user role.
The current user role is several of the family members, and before the media asset library is called, the method also comprises the steps of comparing the priorities preset in the system by the several family members, selecting the media asset library of the family member with the high priority, and calling and displaying the EPG content to the user.
The priority is set when the user image information is recorded.
And calling the content of the media assets in the visitor mode and displaying EPG information to the user when the current user role is not a family member.
The invention provides a flow chart of an optimal embodiment of original picture library entry of a content accurate pushing method based on face recognition, which is shown in figure 2. The method comprises the following specific steps:
step S201: the set top box or the television and other devices collect image information of the home user.
The user image information acquisition is that the set top box or the television and other equipment acquire the face image data of the home user through a camera and other video or picture acquisition modules.
Step S202: and extracting the characteristics of the user image information.
The set top box or the television and other equipment are also used for extracting the characteristics of the collected face image data.
The feature extraction is used for extracting feature point data of the face image.
The facial image feature points include but are not limited to: vertex, glabellar, occipital point, temporal lobe cover, midpoint of hair track, zygomatic point, mandibular corner point, sublabial, submental point, mandibular point, condyloid process outer point, external angle of the eye, center of pupil, orbital lowest point, upper portion of the eyelid, lower portion of the eyelid, orbital highest point, upper point of the eyebrow, nasion point, inferior point of the nasal septum, alar end, nasal bulge point, inferior point of the nasal septum, upper alar end, alar curve, maxillofrontal point, upper labial boundary point, lower labial boundary point, mouth corner, oral point, upper auricular point, lower auricular point, anterior auricular point, posterior auricular point, upper middle auricular point, lower auricular point, upper superior auricular edge midpoint, and tragus point.
Step S203: matching the image information of different family users with the extracted characteristic information, and storing to form an original picture library.
The picture library includes: family member name, role, gender, age, face original picture, face picture structural description information, priority level and associated media asset information.
The family member name, role, gender, age, priority level are manually input by the user.
The related media asset library information is automatically related by the system according to the viewing habits of the users.
Step S204: whether the image information of the family members is collected or not; if not, continuing to acquire the image information of the family members; if so, setting the priority.
Step S205: priority information of different family members is set.
And carrying out priority setting on family members in the original picture library.
The priority setting is that when the face image information of a plurality of family members appears in the picture, the system calls the content of the media asset library corresponding to the family member with the highest priority according to the priority.
The invention provides a flow chart of an optimal embodiment of creating a media asset library corresponding to a user role based on a face recognition content accurate pushing method, which is shown in figure 3. The method comprises the following specific steps:
step S301: when a user watches television program content, the set top box or television and other equipment can automatically acquire the program content information watched by the user.
The program content information refers to program media resource information and program related metadata information.
The program media resource information refers to audio and video content contained in a program.
The program related metadata information includes but is not limited to: program URL address, program name, program type, cast listing, director, tagging information.
And the collected program content information watched by the user and the user information are sent to a media server together.
Step S302: the media server classifies the program content information watched by the user and acquires the label.
And the media server analyzes the watching preference of the user according to the program content information watched by the user, and classifies and acquires the labels.
Step S303: and the media server forms the acquired labels into a label library.
The media server gathers the labels corresponding to the program content information watched by the user to form a content label library.
Step S304: and the media server summarizes the media asset contents associated with the tag library to form a media asset library associated with the user.
And the media server searches related content in the media server based on the content tag library to form a media resource.
The media resources are the media resources which are interested by the user.
The collection of asset media assets is referred to as an asset library.
The media asset library is associated with a specific user, and is used for calling the media asset library when different users watch television.
The invention provides a system architecture diagram of a content accurate pushing system based on face recognition, which is shown in fig. 4. The method specifically comprises the following steps:
media playback device 400, media server 500.
The media playing device 400 generally refers to a device with a media playing function, such as a set-top box or a television, and is used for collecting data and playing media content.
The data acquisition comprises: the method comprises the steps of collecting user picture data and collecting user viewing habits.
The playing of the media content refers to the decoding playing of the audio and video of the program selected by the user.
The media server 500 is used for collecting user data and providing accurate pushing of media asset content.
The collection of user data includes, but is not limited to: user attribute information, and user viewing habit information.
The user attribute information includes but is not limited to: family member name, role, gender, age.
The user viewing habit information refers to data information of programs watched by the user.
The media server 500 generates a media asset library corresponding to the user according to the acquired user data information.
The media server 500 is further configured to call different media asset libraries according to different users, so as to implement accurate pushing of media asset contents.
The invention provides a functional structure block diagram of a media playing device in a system architecture of a content accurate pushing system based on face recognition, as shown in fig. 5. The method specifically comprises the following steps:
a data acquisition unit 401, a face recognition unit 402, an original picture library 403, a media resource acquisition unit 404 and a media resource presentation unit 405.
The data acquisition unit 401 of the media playing device 400 is configured to acquire video or image data of a human face.
The data acquisition unit 401 is typically a camera module.
The face recognition unit 402 of the media playing device 400 performs face recognition by the user and compares the face recognition with the original picture library.
The face recognition unit 402 performs target detection and recognition on the image data acquired by the data acquisition unit 401, and recognizes a face image appearing in the image.
And the step of recognizing the face image also comprises the step of extracting the structural data of the data characteristics of the face image to obtain the structural data of the intelligent analysis result.
The structured data of the face image refers to the extraction of feature points based on the face image.
The machine vision recognition algorithm of the target detection and recognition model is based on a deep convolutional neural network.
The original picture library 403 of the media playing device 400 is used for storing original picture information of a home user.
The original picture library 403 is further configured to store home user attribute information, priority information, and media asset library information corresponding to the home user.
The home subscriber attribute information includes, but is not limited to: family member name, role, gender, age.
The media asset library information corresponds to the family members one by one.
The media asset obtaining unit 404 of the media playing device 400 is configured to obtain media asset data.
The media asset acquisition unit 404 acquires media asset data from the media server.
The media asset presenting unit 405 of the media playing device 400 is configured to present media asset data.
The media asset presenting unit 405 is configured to decode and display the media asset data acquired by the media asset acquiring unit 404.
The invention provides a functional structure block diagram of a media server in a system architecture of a content accurate pushing system based on face recognition, as shown in fig. 6. The method specifically comprises the following steps:
a user information acquisition unit 501, a media asset classification unit 502, a media asset library 503 and a media asset pushing unit 504.
The user information collecting unit 501 of the media server 500 is configured to collect user viewing habit data.
The user viewing habit data comprises: user information and metadata information for the program.
The user information includes but is not limited to: family member name, role, gender, age.
The metadata information of the program includes, but is not limited to: program URL address, program name, program type, cast listing, director, tagging information.
The asset classification unit 502 of the media server 500 is configured to form an asset classification corresponding to a user according to a viewing habit of the user.
The media asset classifying unit 502 classifies and labels the program content information viewed by the user.
The media asset classification unit 502 is further configured to aggregate tags corresponding to program content information watched by a user to form a content tag library.
The media asset classification unit 502 searches for related content in the media asset library 503 according to the content tag library to form media assets.
The media resources are associated with specific users, and are used for calling the media resource library when different users watch television.
The media asset library 503 of the media server 500 is configured to store media asset audio/video files and metadata files.
The metadata is description information of the program.
The media asset pushing unit 504 of the media server 500 is configured to push media asset data to a user.
The media asset pushing unit 504 is configured to call different media asset libraries to send to the terminal user when different users watch the television.
The content accurate pushing method and device based on face recognition can apply a machine vision face recognition technology, realize accurate pushing of content and greatly improve user experience.
It should be understood that the invention is not limited to the embodiments described above, but that modifications and variations can be made by one skilled in the art in light of the above teachings, and all such modifications and variations are intended to be within the scope of the invention as defined by the appended claims.

Claims (10)

1. A content accurate pushing method based on face recognition is characterized by comprising the following steps:
starting up and collecting video or picture data;
intelligently analyzing the collected video or picture data and obtaining intelligent analysis result structured data;
matching the intelligent analysis result with an original picture library to obtain the current user role;
and calling the corresponding media asset library according to the current user role, and displaying the media asset library to the user.
2. The content accurate pushing method based on face recognition as claimed in claim 1, wherein the starting up and capturing video or picture data specifically comprises:
the starting-up refers to the starting-up of equipment with a media playing function, such as a set-top box or a television;
the set top box or the television is provided with a video or picture data acquisition module and is used for acquiring video or picture data;
the video or picture acquisition module comprises but is not limited to a camera or camera-like equipment with a video or picture acquisition function;
the acquisition of the video or the picture refers to that the video or picture acquisition module acquires video or picture data when the set top box or the television and other devices are started.
3. The content accurate pushing method based on face recognition as claimed in claim 1, wherein the intelligent analysis of the collected video or picture data to obtain the structured data of the intelligent analysis result specifically comprises:
the acquired data is video data, and frame extraction processing needs to be carried out on the video data to obtain picture data;
the intelligent analysis refers to the intelligent analysis of the picture data obtained by frame extraction or the collected picture data;
the intelligent analysis is to perform target detection and identification on the picture data and identify a face image appearing in the picture;
the face image recognition also comprises the step of extracting the structural data of the face image data characteristics to obtain the structural data of the intelligent analysis result;
the machine vision recognition algorithm of the target detection and recognition model is based on a deep convolutional neural network.
4. The method for accurately pushing content based on face recognition as claimed in claim 1, wherein the matching of the enabled analysis result with the original picture library to obtain the current user role specifically comprises:
the intelligent analysis result refers to the structural data description of the human face features;
the original picture library is a picture library of family members input by a user in advance;
the image is matched with the original image library, and is actually matched with the structural data description of the face features in the original image library;
the matching outputs matching result information according to the similarity;
the matching result is one or more of the family members, or not the family members.
5. The method for accurately pushing content based on face recognition according to claim 1, wherein the invoking of the corresponding media asset library according to the current user role and the displaying to the user specifically comprise:
the current user role is a certain one of family members, a media asset library corresponding to the member is called, and EPG content is displayed to the user;
the current user role is several of the family members, and before the media asset library is called, the method also comprises the steps of comparing the priorities preset in the system by the several family members, selecting the media asset library of the family member with the high priority, and calling and displaying the EPG content to the user;
and calling the content of the media assets in the visitor mode and displaying EPG information to the user when the current user role is not a family member.
6. The method according to claim 4, wherein the prior entry of the original picture library specifically comprises:
the set top box or the television and other equipment acquire image information of the home user;
extracting the characteristics of the user image information;
matching and storing the image information of different family users with the extracted characteristic information to form an original picture library;
after the family member picture library information is input, priority information of different family members is required to be set.
7. The method of claim 5, wherein the media asset library corresponding to the user role specifically comprises:
when the user watches the television program content, the set top box or the television and other equipment can automatically acquire the program content information watched by the user;
the media server classifies the program content information watched by the user and acquires the label;
the media server forms the acquired labels into a label library;
and the media server summarizes the media asset contents associated with the tag library to form a media asset library associated with the user.
8. The utility model provides an accurate push system of content based on face identification which characterized in that, the system specifically includes:
the set-top box or the television and other equipment with the media playing function are used for collecting data and playing media contents;
and the media server is used for collecting user data and providing accurate pushing of media asset content.
9. The system for accurately pushing content based on face recognition as claimed in claim 8, wherein the set-top box or television device specifically comprises:
the data acquisition unit is used for acquiring video or image data of a human face;
the face recognition unit is used for carrying out face recognition by a user and comparing the face recognition with an original picture library;
the original picture library is used for storing original picture information of a home user;
the media resource acquisition unit is used for acquiring media resource data;
and the media asset presenting unit is used for presenting the media asset data.
10. The system for accurately pushing content based on face recognition as claimed in claim 8, wherein the media server specifically comprises:
the user information acquisition unit is used for acquiring user viewing habit data;
the media asset classification unit is used for forming media asset classifications corresponding to the users according to the viewing habits of the users;
the media asset library is used for storing media asset audio and video files and metadata files;
and the media asset pushing unit is used for pushing media asset data to the user.
CN202010569981.7A 2020-06-21 2020-06-21 Content accurate pushing method and system based on face recognition Pending CN111741364A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010569981.7A CN111741364A (en) 2020-06-21 2020-06-21 Content accurate pushing method and system based on face recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010569981.7A CN111741364A (en) 2020-06-21 2020-06-21 Content accurate pushing method and system based on face recognition

Publications (1)

Publication Number Publication Date
CN111741364A true CN111741364A (en) 2020-10-02

Family

ID=72651972

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010569981.7A Pending CN111741364A (en) 2020-06-21 2020-06-21 Content accurate pushing method and system based on face recognition

Country Status (1)

Country Link
CN (1) CN111741364A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112434661A (en) * 2020-12-11 2021-03-02 四川长虹电器股份有限公司 Face recognition and classification registration method, computer equipment and storage medium
CN113271483A (en) * 2021-05-19 2021-08-17 中山亿联智能科技有限公司 Intelligent set top box with face recognition function
CN115080788A (en) * 2022-08-11 2022-09-20 小米汽车科技有限公司 Music pushing method and device, storage medium and vehicle

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103873941A (en) * 2012-12-17 2014-06-18 联想(北京)有限公司 Display method and electronic equipment
CN104320708A (en) * 2014-10-14 2015-01-28 小米科技有限责任公司 User right handling method and device of smart television
CN105426850A (en) * 2015-11-23 2016-03-23 深圳市商汤科技有限公司 Human face identification based related information pushing device and method
CN106303699A (en) * 2016-08-24 2017-01-04 三星电子（中国）研发中心 For the method and apparatus playing TV programme
CN107948737A (en) * 2017-11-17 2018-04-20 上海与德科技有限公司 The recommendation method and device of TV programme

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103873941A (en) * 2012-12-17 2014-06-18 联想(北京)有限公司 Display method and electronic equipment
CN104320708A (en) * 2014-10-14 2015-01-28 小米科技有限责任公司 User right handling method and device of smart television
CN105426850A (en) * 2015-11-23 2016-03-23 深圳市商汤科技有限公司 Human face identification based related information pushing device and method
CN106303699A (en) * 2016-08-24 2017-01-04 三星电子（中国）研发中心 For the method and apparatus playing TV programme
CN107948737A (en) * 2017-11-17 2018-04-20 上海与德科技有限公司 The recommendation method and device of TV programme

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112434661A (en) * 2020-12-11 2021-03-02 四川长虹电器股份有限公司 Face recognition and classification registration method, computer equipment and storage medium
CN113271483A (en) * 2021-05-19 2021-08-17 中山亿联智能科技有限公司 Intelligent set top box with face recognition function
CN115080788A (en) * 2022-08-11 2022-09-20 小米汽车科技有限公司 Music pushing method and device, storage medium and vehicle

Similar Documents

Publication Publication Date Title
US20210084369A1 (en) Methods for identifying video segments and displaying contextually targeted content on a connected television
US10271098B2 (en) Methods for identifying video segments and displaying contextually targeted content on a connected television
CN110933490B (en) Automatic adjustment method for picture quality and tone quality, smart television and storage medium
CN111741364A (en) Content accurate pushing method and system based on face recognition
CN109089157B (en) Video picture cutting method, display device and device
US9176987B1 (en) Automatic face annotation method and system
US9003445B1 (en) Context sensitive thumbnail generation
EP2541963A2 (en) Method for identifying video segments and displaying contextually targeted content on a connected television
US20170150210A1 (en) Devices, systems, methods, and media for detecting, indexing, and comparing video signals from a video display in a background scene using a camera-enabled device
EP3031205A1 (en) Systems and methods for providing synchronized content
JP5637930B2 (en) Interest section detection device, viewer interest information presentation device, and interest section detection program
JP2002521977A (en) Apparatus and method for locating commercials located within a video data stream
CN112004108B (en) Live video recording processing method and device, intelligent terminal and storage medium
CN112291589B (en) Method and device for detecting structure of video file
KR20150083355A (en) Augmented media service providing method, apparatus thereof, and system thereof
CN105979366A (en) Smart television and content recommending method and content recommending device thereof
JP2007274246A (en) Receiver and program
CN107948737A (en) The recommendation method and device of TV programme
KR20170102570A (en) Facilitating television based interaction with social networking tools
CN111444822B (en) Object recognition method and device, storage medium and electronic device
CN115396705B (en) Screen operation verification method, platform and system
CN110990625A (en) Movie recommendation method and device
CN113423014B (en) Playing information pushing method and device, terminal equipment and storage medium
CN112822539A (en) Information display method, device, server and storage medium
CN109151599B (en) Video processing method and device

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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20201002