Video recommendation method and device based on face recognition, television

Abstract

A video recommendation method and device for a television based on face recognition includes the steps of: acquiring a face image containing at least one user's face (S1); acquiring a facial feature set of the user from the face image based on a face recognition algorithm (S2); comparing the facial feature set with a facial feature set in an archive memory (S3); displaying a first video recommendation set if at least one user's facial feature set matches the facial feature set in the archive memory; and displaying a second video recommendation set if the respective facial feature sets do not match the facial feature set in the archive memory (S4). This method recommends videos to users, reduces the number of operational steps required for the user to select a video to watch, and improves user convenience.

Description

[Technical field]

[0001] The present invention relates to the technical field of smart televisions, and in particular to a video recommendation method and device based on face recognition, a television, and a storage medium. [Background technology]

[0002] With the progress of science, people's living standards are improving, and people are paying more and more attention to life experience. Everyone has their own hobbies, and different people like different movies. Currently, when a TV recommends videos to a user, the user needs to log in to his / her account on the TV to get the recommended content, which makes the operation flow complicated and makes it difficult for users to use.

[0003] Therefore, there is room for improvement in the prior art. Summary of the Invention [Problem to be solved by the invention]

[0004] The problem that the present invention aims to solve is to provide a video recommendation method and device, television, and storage medium based on facial recognition in order to reduce the operational steps required for a user to select a video to watch and improve user convenience. [Means for solving the problem]

[0005] In a first aspect, an embodiment of the present invention comprises: acquiring a face image including a face of at least one user; obtaining a set of facial features of a user from the facial image based on a facial recognition algorithm; matching the set of facial features with facial feature sets in an archive memory; displaying a first video recommendation set if at least one of the users' facial feature sets matches a facial feature set in the archive memory; and displaying a second video recommendation set if the respective facial feature set does not match the facial feature set in the archive memory.

[0006] In one embodiment, before acquiring the facial image, The method further includes the step of constructing the archive memory in advance.

[0007] In one embodiment, the step of constructing the archive memory in advance includes: Collecting a facial image of a user; obtaining a set of facial features of the user from a facial image of the user based on a facial recognition algorithm; and storing a set of facial features of the user corresponding to the user's account in the archival memory.

[0008] In one embodiment, the step of obtaining a set of facial features of the user from the face image based on the face recognition algorithm includes: processing the facial image to obtain a processed facial image; performing facial feature recognition on the processed facial image to obtain facial features of a user; and filtering the user's facial features to obtain a set of the user's facial features including at least the user's sense features and facial contour features.

[0009] In one embodiment, processing the facial image to obtain a processed facial image includes: converting the acquired face image from an analog signal to a digital signal to obtain a first image; binarizing the first image to obtain a second image; smoothing the second image to obtain a third image; and performing a transformation on the third image to correct for systematic errors in the third image to obtain the processed face image.

[0010] In one embodiment, if the facial feature set of the at least one user matches a facial feature set in the archive memory, the step of displaying a first set of video recommendations includes: and displaying the first video recommendation set if a degree of match between the facial feature set of at least one user and the facial feature set in the archive memory is greater than or equal to a threshold value.

[0011] In one embodiment, if the respective facial feature sets do not match the facial feature sets in the archive memory, the step of displaying a second set of video recommendations comprises: If a degree of match between the facial feature set of the at least one user and the facial feature set in the archive memory is below a threshold, displaying a second set of video recommendations.

[0012] In one embodiment, the first video recommendation set includes a first video recommendation subset and a second video recommendation subset, and when the facial feature set of the at least one user matches a facial feature set in the archive memory, the step of displaying the first video recommendation set includes: When the facial feature set of only one user among the facial feature sets of the at least one user matches the facial feature set in the archive memory, displaying a first video recommendation subset corresponding to the user; When facial feature sets of multiple users among the facial feature sets of the at least one user match the facial feature sets in the archive memory, displaying a second video recommendation subset corresponding to the facial feature sets of the multiple users.

[0013] In one embodiment, the first video recommendation subset corresponding to the user is a user preferred video set corresponding to the user.

[0014] In one embodiment, the second video recommendation subset corresponding to the facial feature sets of the plurality of users is an intersection of the plurality of user preferred video sets corresponding to the plurality of users.

[0015] In one embodiment, the second video recommendation set is a crowd-pleasing video set.

[0016] In one embodiment, before acquiring the facial image, The method further includes determining a user preferred video set based on the viewing history of the user's account.

[0017] In one embodiment, if the respective facial feature set does not match the facial feature set in the archive memory, after displaying the second video recommendation set, establishing an account for said at least one user; storing the facial feature set of the at least one user in the archival memory and correspondingly associating the facial feature set of the user with an account of the user.

[0018] In a second aspect, an embodiment of the invention comprises: a first acquisition module for acquiring a face image including a face of at least one user; a second acquisition module for acquiring a set of facial features of the user from the face image based on a face recognition algorithm; a matching module for matching the set of facial features with the set of facial features in the archive memory; and a video recommendation module for displaying a first video recommendation set when at least one user's facial feature set matches a facial feature set in the archive memory, and for displaying a second video recommendation set when each of the facial feature sets does not match the facial feature set in the archive memory.

[0019] In a third aspect, an embodiment of the present invention comprises a memory having a computer program stored therein, the computer program being executed, acquiring a face image including a face of at least one user; obtaining a set of facial features of the user from the facial image based on a facial recognition algorithm; matching the set of facial features with sets of facial features in an archive memory; displaying a first video recommendation set if at least one of the users' facial feature sets matches a facial feature set in the archive memory; and if the respective set of facial features does not match the set of facial features in the archive memory, displaying a second set of video recommendations.

[0020] In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium having stored thereon a computer program, the computer program, when executed by a processor, comprising: acquiring a face image including a face of at least one user; obtaining a set of facial features of the user from the facial image based on a facial recognition algorithm; matching the set of facial features with sets of facial features in an archive memory; displaying a first video recommendation set if at least one of the users' facial feature sets matches a facial feature set in the archive memory; The present invention further provides a computer-readable storage medium implementing the step of: displaying a second set of video recommendations if the respective set of facial features does not match the set of facial features in the archive memory. Effect of the Invention

[0021] The embodiment of the present invention has the following advantages over the prior art. According to the method of the embodiment of the present invention, first, a face image including at least one user's face is obtained, then a facial feature set of the user is obtained from the face image based on a facial recognition algorithm, and finally, the facial feature set is compared with the facial feature set in the archive memory, and if the facial feature set of at least one user matches the facial feature set in the archive memory, a first video recommendation set is displayed, and if each facial feature set does not match the facial feature set in the archive memory, a second video recommendation set is displayed. This method recommends videos to the user, reduces the operation steps of the user selecting the video to watch, and improves the user's convenience. [Brief description of the drawings]

[0022] In the following, in order to more clearly explain the embodiments of the present invention or the technical means in the prior art, drawings that need to be used in the description of the embodiments or the prior art will be briefly described. However, the drawings in the following description are only some embodiments of the present invention, and it is obvious that those skilled in the art can derive other drawings from these drawings without creative efforts. [Figure 1] FIG. 1 is a flowchart of a video recommendation method based on face recognition in an embodiment of the present invention. [Diagram 2] FIG. 2 is a structural schematic diagram of a video recommendation device based on face recognition in an embodiment of the present invention. [Diagram 3] FIG. 3 is a diagram showing the internal structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] In order to allow those skilled in the art to better understand the technical means of the present invention, the technical means in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention. It is clear that the described embodiments are only a part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments that can be obtained by those skilled in the art without creative labor belong to the protection scope of the present invention.

[0024] Research has shown that in the prior art, when a TV recommends videos to a user, the user needs to log in to his or her account on the TV to obtain the recommended content, which makes the operation flow complicated and results in a poor user experience.

[0025] To solve the above problems, in an embodiment of the present invention, a TV can collect images of a user using a camera, and then recognize the user from the face of the collected images, and recommend videos that the user can enjoy based on the recognition result in combination with big data. According to the video recommendation method in the embodiment of the present invention, a smart TV can display the user's favorite content through recommendations, allowing the user to easily find movies he or she wants to watch, improving the user experience, reducing the user's operation flow, making content recommendations to the user more accurate, and facilitating user convenience. And, smart recommendation provides accurate recommendation services in real time for various scenarios by investigating the user's behavior and service features in detail, and quickly improves the user's activity level and click conversion rate.

[0026] Various non-limiting embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0027] An embodiment of the present invention provides a video recommendation method based on face recognition, and as shown in FIG. 1, the method includes steps S1 to S4.

[0028] S1: Obtain a face image that includes at least one user's face.

[0029] In an embodiment of the present invention, a user's face image or video stream can be acquired by a camera in a smart TV. Multiple users may be watching the smart TV, and multiple users may be present in the face image.

[0030] In a preferred embodiment of the present invention, before starting to use the video recommendation function of the present invention, the archive memory needs to be pre-constructed, that is, before step S1, step S0 is included.

[0031] S0: The archive memory is constructed in advance.

[0032] In an embodiment of the present invention, a set of facial features of a user is collected and stored in an archive memory, specifically, the process of constructing the archive memory, ie, step S0, includes steps S01 to S03.

[0033] S01: Collect a face image of a user.

[0034] In an embodiment of the present invention, a user's face image or video stream is collected by a camera in a smart TV, and the user's face image is further obtained from the video stream collected by the camera.

[0035] S02: Obtain a set of facial features of the user from a facial image of the user.

[0036] In an embodiment of the present invention, a plurality of facial features are obtained from a user's face image based on a face recognition algorithm, and then the user's sense features and contour features are selected from the obtained plurality of facial features to form a facial feature set.

[0037] S03: Store the user's facial feature set corresponding to the user's account in the archive memory.

[0038] In an embodiment of the present invention, the facial feature set and the features in the facial feature set are stored in an archival memory in the form of a binary field. The archival memory is stored in a cloud server.

[0039] For example, before a user Mr. Liu uses the video recommendation function of a smart TV, an archive memory of a person who uses the smart TV is constructed in advance, and facial feature sets of Mr. Liu and his wife Ms. Wang are required. Specifically, face images of Mr. Liu and Mr. Wang are collected respectively, and face feature sets of Mr. Liu and Mr. Wang are obtained from the face images of Mr. Liu and Mr. Wang respectively based on a face recognition algorithm, and then the face feature sets of Mr. Liu and Mr. Wang are stored in the archive memory respectively, where the face feature set of Mr. Liu corresponds to Mr. Liu's account, and the account of Mr. Wang corresponds to Mr. Wang. Through the above steps, when a user uses a smart TV, the user can be recognized and automatically logged in, and subsequently videos of the user's preference type can be recommended to the user.

[0040] S2: Obtain a set of user's facial features from the face image based on a face recognition algorithm.

[0041] In an embodiment of the present invention, a set of facial features is obtained from the face image based on a face recognition algorithm, and face recognition is a biometric technology that performs identity recognition based on a person's facial feature information. A series of related technologies that collect face images or video streams using a camera or webcam, automatically detect and track facial features in the face images, and perform face recognition on the detected faces are generally called image recognition, face recognition, etc. When there are multiple face images, a set of facial features corresponding to each of the multiple face images is obtained based on a face recognition algorithm.

[0042] In a preferred embodiment of the present invention, step S2 includes steps S21 to S23.

[0043] S21: Process the face image to obtain a processed face image.

[0044] In an embodiment of the present invention, the facial image is processed, including performing analog-to-digital conversion, binarization, and smoothing on the facial image.

[0045] Specifically, step S21 includes steps S211 to S213.

[0046] S211: Convert the collected face image from an analog signal to a digital signal to obtain a first image.

[0047] In an embodiment of the present invention, the facial image is pre-processed and the analog signal received by the camera is converted to a digital signal by an A / D converter.

[0048] S212: The first image is binarized to obtain a second image.

[0049] In the embodiment of the present invention, the grayscale value of the pixel dots of the image is set to 0 or 255, which is the process of showing the whole image as a noticeable black and white effect.

[0050] S213: The second image is smoothed to obtain a third image.

[0051] In an embodiment of the present invention, due to the influence of the sensor, the atmosphere, etc., a remote sensing image may have a region with a large change in brightness or may have bright spots (also called noise). In order to suppress such noise, a processing method for smoothing the brightness of an image is image smoothing. Image smoothing is actually a low-pass filter process, and the smoothing process blurs the edges of the image. In order to suppress noise in the target image while retaining the detailed features of the image as much as possible, the noise is reduced by filtering.

[0052] S214: A transformation for correcting a systematic error in the third image is performed on the third image to obtain the processed face image.

[0053] In an embodiment of the present invention, performing transformation on the third image in step S214 includes at least performing processing on the third image to correct system errors of the image acquisition system and random errors of the equipment position (causes of the imaging angle, perspective relationship, or the lens itself) by performing geometric transformations such as translation, transposition, mirroring, rotation, and scaling on the third image.

[0054] The collected facial images are processed to remove useless information in the acquired facial images and highlight useful actual information in the acquired facial images, so that the processed facial images can easily recognize the user's face in subsequent steps, and the recognition effect is improved.

[0055] In an embodiment of the present invention, the face recognition algorithm is for recognizing and extracting facial feature points to obtain a set of facial features of a user. For example, the open source OpenCV algorithm is used to extract facial feature points. First, a face is detected using the face detector of OpenCV to obtain facial key points, that is, the position points of the eyes, nose, mouth, etc. can be obtained. These key points and their combination relationships are facial feature points.

[0056] In an embodiment of the present invention, the steps of the face recognition algorithm include: 1. using OpenCV Haar face detector or lbp face detector to detect faces in the acquired face image; 2. creating an object of Facemark class, where Facemark uses a smart pointer (PTR) in OpenCV; 3. loading a keypoint detector (lbfmodel.yaml), which is trained on thousands of face images with keypoint labels; 4. running the face detector on the acquired face image, the output of the face detector is a container (vector) containing one or more rectangles, i.e., there may be one or more faces in the face image; 5. extracting a face ROI in the original image based on the frame of the face rectangle, and using a facial landmark detector to detect the face ROI, and obtaining multiple keypoints for each face, which are stored in a set; and 6. drawing and displaying the face image based on the obtained keypoints.

[0057] S22: Perform facial feature recognition on the processed face image to obtain the facial features of the user.

[0058] For example, when Mr. Liu and his wife Ms. Wang watch television together, a television camera collects facial images, which are processed to obtain processed facial images, and then facial recognition is performed on the processed facial images to obtain facial features of the two users, Mr. Liu and Ms. Wang.

[0059] S23: Filter the user's facial features to obtain a user's facial feature set including at least the user's five senses features and facial contour features.

[0060] In an embodiment of the present invention, when there are multiple users' faces in the face image, facial feature sets of multiple users are obtained.

[0061] For example, a 64x64 image can obtain 4096 pixel dot data, and the points in the image are divided into different subsets, and these subsets often belong to isolated points, continuous curves, or continuous regions. The location of the user's moles can be characterized, the size of the eyes can be characterized, the position of the eyes relative to the nose can be characterized, the shape of the mouth, etc. The more features, the higher the accuracy of the recognition.

[0062] S3: Match the facial feature set with the facial feature set in the archive memory.

[0063] In an embodiment of the present invention, facial feature sets of a plurality of users are stored in advance in an archive memory, and the facial feature set of each of the plurality of users is compared with the facial feature set in the archive memory. When comparing, the degree of match between each facial feature set and the facial feature set of each user previously stored in the archive memory is determined. Each degree of match is compared with a threshold, and if each degree of match is smaller than the threshold, it is determined that each facial feature set does not match the facial feature set in the archive memory. If any degree of match is greater than the threshold, it is determined that at least one user's facial feature set matches the facial feature set in the archive memory.

[0064] When there is only one user, if the degree of matching between the user's facial feature set and the facial feature set in the archive memory is equal to or greater than a threshold, matching is successful and the user's facial feature set matches the facial feature set in the archive memory, and if the degree of matching between the user's facial feature set and the facial feature set in the archive memory is less than the threshold, matching is unsuccessful and the user's facial feature set does not match the facial feature set in the archive memory.When there are multiple users, if the degree of matching between one user's facial feature set and the facial feature set in the archive memory is equal to or greater than a threshold, matching is successful, and if the degree of matching between all users' facial feature sets and the facial feature sets in the archive memory is less than the threshold, matching is unsuccessful.

[0065] For example, a set of facial features of a user obtained from a face image includes a1 and a2, and facial feature sets of multiple users pre-stored in an archive memory include b1 and b2, and when comparing, the degree of match between a1 and b1 is determined, the degree of match between a1 and b2 is determined, the degree of match between a2 and b1 is determined, and the degree of match between a2 and b2 is determined.

[0066] S4: If at least one user's facial feature set matches the facial feature set in the archive memory, display a first video recommendation set, and if each facial feature set does not match the facial feature set in the archive memory, display a second video recommendation set.

[0067] In an embodiment of the present invention, the first video recommendation set is a user preference video set, and the second video recommendation set is a public preference video set.

[0068] Specifically, step S4 includes steps S41 and S42.

[0069] S41: If the degree of match between at least one user's facial feature set and the facial feature set in the archive memory is equal to or greater than a threshold, a first video recommendation set is displayed.

[0070] S42: If the degree of match between the facial feature set of the at least one user and the facial feature set in the archive memory is less than a threshold, display a second video recommendation set.

[0071] In the present invention, a video recommendation set to be displayed is determined based on the degree of matching between at least one user's facial feature set and the facial feature set in the archive memory. At least one user's facial feature set is a1, and the pre-stored facial feature sets of multiple users include b1 and b2. If the degree of matching between a1 and b1 is greater than a threshold, a first video recommendation set corresponding to b1 is broadcast; if the degree of matching between a1 and b1 is less than the threshold and the degree of matching between a1 and b2 is also less than the threshold, a second video recommendation set is displayed.

[0072] In a preferred aspect of the embodiment of the present invention, the first video recommendation set includes a first video recommendation subset and a second video recommendation subset, and if the facial feature set of the at least one user matches a facial feature set in the archive memory, the step of displaying the first video recommendation set includes: When the facial feature set of only one user among the facial feature sets of the at least one user matches the facial feature set in the archive memory, displaying a first video recommendation subset corresponding to the user; When facial feature sets of multiple users among the facial feature sets of the at least one user match the facial feature sets in the archive memory, displaying a second video recommendation subset corresponding to the facial feature sets of the multiple users.

[0073] In an embodiment of the present invention, the first video recommendation subset corresponding to the user is a user preference video set corresponding to the user, which is a common set of a plurality of user preference video sets corresponding to the plurality of users. If the acquired user faces are multiple and the multiple facial feature sets match the facial feature set in the archive memory, a second video recommendation subset is displayed.

[0074] In an embodiment of the present invention, when a facial feature set of one user matches a facial feature set in the archive memory, a first video recommendation subset is displayed, the first video recommendation subset being the user's preferred video set; when facial feature sets of multiple users match the facial feature sets in the archive memory, a second video recommendation subset is displayed, the second video recommendation subset being a common set of the multiple user preferred video sets.

[0075] In an embodiment of the present invention, before obtaining the face image, the method further includes determining a user's preferred video set based on a viewing history in a user's account, specifically, determining the user's preferred video set based on the viewing history in a user's account in combination with a big data algorithm.

[0076] and establishing an account for the at least one user after displaying a second video recommendation set if each facial feature set does not match the facial feature set in the archive memory, and storing the facial feature set of the at least one user in the archive memory and correspondingly associating the facial feature set of the user with the account of the user. Thus, when the at least one user next uses, the video recommendation method based on face recognition according to the present invention can display the corresponding first video recommendation set to the at least one user.

[0077] According to the video recommendation method of the embodiment of the present invention, the TV can display the user's favorite content through smart recommendation, so that the user can easily find the movie he or she wants to watch, improve the user experience, reduce the user's operation flow, more accurately recommend content to the user, and facilitate the user's convenience. And, the smart recommendation can provide accurate recommendation services in real time for various scenarios by investigating the user's behavior and service features in detail, and quickly improve the user's activity and click conversion rate.

[0078] An embodiment of the present invention provides a video recommendation device based on face recognition, as shown in FIG. 2, the device includes: a first acquisition module 20 for acquiring a face image including a face of at least one user; a second acquisition module 22 for acquiring a set of facial features of the user from the face image based on a face recognition algorithm; a matching module 24 for matching the set of facial features with facial feature sets in an archive memory; and a video recommendation module 26 for displaying a first video recommendation set when at least one user's facial feature set matches the facial feature set in the archive memory, and for displaying a second video recommendation set when the respective facial feature set does not match the facial feature set in the archive memory.

[0079] According to the video recommendation method of the embodiment of the present invention, the TV can display the user's favorite content through smart recommendation, so that the user can easily find the movie he or she wants to watch, improve the user experience, reduce the user's operation flow, more accurately recommend content to the user, and facilitate the user's convenience. And, the smart recommendation can provide accurate recommendation services in real time for various scenarios by investigating the user's behavior and service features in detail, and quickly improve the user's activity and click conversion rate.

[0080] In one embodiment, the present invention provides a computer device, which may be a terminal, and its internal configuration is shown in FIG. 3. The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the computer device is for providing calculation and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. An operating system and a computer program are stored in the non-volatile storage medium. The internal memory provides an environment for the execution of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is for connecting and communicating with an external terminal via a network. When the computer program is executed by the processor, the method for generating a natural language model is realized. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered by the display screen, a button, a trackball, or a touch panel provided on the housing of the computer device, or an external keyboard, touch panel, mouse, etc.

[0081] Those skilled in the art will understand that the configuration shown in FIG. 3 is merely a block diagram of a portion of the structure related to the technical means of the present invention, and does not limit the computer device to which the technical means of the present invention is applied; a specific computer device may include more or fewer components than those shown in the configuration, may combine some of the components, or may have different components.

[0082] An embodiment of the present invention includes a memory in which a computer program is stored, and, when executing the computer program, acquiring a face image including a face of at least one user; obtaining a set of facial features of the user from the facial image based on a facial recognition algorithm; matching the set of facial features with sets of facial features in an archive memory; displaying a first video recommendation set if at least one of the users' facial feature sets matches a facial feature set in the archive memory; and a processor for implementing the step of displaying a second set of video recommendations if the respective set of facial features does not match the set of facial features in the archive memory.

[0083] An embodiment of the present invention is a computer-readable storage medium having a computer program stored thereon, the computer program, when executed by a processor, comprising: acquiring a face image including a face of at least one user; obtaining a set of facial features of the user from the facial image based on a facial recognition algorithm; matching the set of facial features with sets of facial features in an archive memory; displaying a first video recommendation set if at least one of the users' facial feature sets matches a facial feature set in the archive memory; The present invention further provides a computer-readable storage medium implementing the step of: displaying a second set of video recommendations if the respective set of facial features does not match the set of facial features in the archive memory.

[0084] The technical features of the above embodiments can be combined in any manner. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as no contradiction occurs in the combination of these technical features, they should be considered within the scope of the present specification.

[0085] The above examples merely show some embodiments of the present invention, and the description is specific and detailed, but should not be understood as limiting the scope of the present invention. It should be noted that a person skilled in the art can make various modifications and improvements without departing from the spirit of the present invention, and all of these are included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be in accordance with the scope of the attached claims.

[0086] This invention claims priority to a Chinese patent application filed with the State Intellectual Property Office of the People's Republic of China on December 30, 2019, bearing application number 201911391967.6 and titled "Video recommendation method and apparatus based on face recognition, television," the entire contents of which are incorporated herein by reference.

Claims

1. acquiring a face image including a face of at least one user; obtaining a set of facial features of a user from the facial image based on a facial recognition algorithm; matching the set of facial features with facial feature sets in an archive memory; displaying a first set of video recommendations if at least one of the users' facial feature sets matches a facial feature set in the archive memory; displaying a second video recommendation set if the respective facial feature set does not match the facial feature set in the archive memory; establishing an account for a user whose facial feature set does not match a facial feature set in the archive memory after displaying the second video recommendation set, and storing the corresponding user's facial feature set in the archive memory in association with the user's account; A video recommendation method based on face recognition, comprising:

2. Before acquiring the face image, The method for recommending video based on face recognition according to claim 1 , further comprising the step of constructing the archive memory in advance.

3. The step of constructing an archive memory in advance includes: Collecting a facial image of a user; obtaining a set of facial features of the user from a facial image of the user based on a facial recognition algorithm; and storing the user's facial feature set corresponding to the user's account in the archive memory.

4. The step of obtaining a set of facial features of the user from the face image based on the face recognition algorithm includes: processing the facial image to obtain a processed facial image; performing facial feature recognition on the processed facial image to obtain facial features of a user; The method of claim 1 , further comprising: filtering the user's facial features to obtain a set of the user's facial features including at least the user's five senses features and facial contour features.

5. Processing the facial image to obtain a processed facial image, converting the acquired face image from an analog signal to a digital signal to obtain a first image; binarizing the first image to obtain a second image; smoothing the second image to obtain a third image; 5. The method of claim 4, further comprising: performing a transformation on the third image to correct for systematic errors in the third image to obtain the processed face image.

6. If the facial feature set of the at least one user matches a facial feature set in the archive memory, displaying a first video recommendation set includes: The method for recommending videos based on face recognition according to claim 1, further comprising: displaying a first video recommendation set if a degree of match between the facial feature set of at least one user and the facial feature set in the archive memory is equal to or greater than a threshold value.

7. If the respective facial feature sets do not match the facial feature sets in the archive memory, displaying a second video recommendation set includes: The method for recommending videos based on face recognition according to claim 1 , further comprising: displaying a second video recommendation set if the degree of match between the facial feature set of the at least one user and the facial feature set in the archive memory is less than a threshold value.

8. The first video recommendation set includes a first video recommendation subset and a second video recommendation subset, and if the facial feature set of the at least one user matches a facial feature set in the archive memory, the step of displaying the first video recommendation set includes: displaying a first video recommendation subset corresponding to a user when the facial feature set of only one user among the facial feature sets of the at least one user matches the facial feature set in the archive memory; 2. The method of claim 1, further comprising: displaying a second video recommendation subset corresponding to the facial feature sets of the plurality of users among the facial feature sets of the at least one user when the facial feature sets of the plurality of users match the facial feature sets in the archive memory.

9. The method of claim 8 , wherein the first video recommendation subset corresponding to the user is a user preference video set corresponding to the user.

10. The method of claim 8 , wherein the second video recommendation subset corresponding to the facial feature sets of the plurality of users is a common set of the plurality of user preference video sets corresponding to the plurality of users.

11. The method of claim 1 , wherein the second video recommendation set is a popular video set.

12. Before acquiring the face image, The face recognition based video recommendation method of claim 1 , further comprising: determining a user preferred video set based on a viewing history in a user's account.

13. A television comprising a memory in which a computer program is stored and a processor which, when executing said computer program, implements the steps of the method according to any one of claims 1 to 12.

14. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 12.

Images