Material video duplication checking method and device, electronic equipment and computer readable medium

Abstract

Embodiments of the present disclosure disclose a material video duplicate detection method and device, electronic equipment and a computer readable medium. A specific embodiment of the method comprises: receiving a material video as a video to be detected; selecting each comparison material video from a material video library; analyzing the video to be detected to obtain a frame sequence of the video to be detected; for each comparison material video, the following steps are performed: generating each key frame feature information of the video to be detected; determining the comparison material video as a similar material video in response to determining that the comparison material video meets the key frame matching condition according to each key frame feature information of the video to be detected; performing duplicate detection processing on the similar material video based on each target key frame of the video to be detected to determine material video duplicate detection information; and sending each material duplicate detection information to a user terminal for displaying each material duplicate detection information by the user terminal. The embodiment can reduce the resource consumption and the time consumption of video duplicate detection.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of computer, and specifically to a material video duplicate detection method and device, electronic equipment and computer readable medium. BACKGROUND

[0002] Video duplicate detection is a common video matching technology, which is used to find target videos with high similarity or containing duplicate video segments from multiple videos. At present, when performing video duplicate detection, the commonly used method is to extract video or video frame image feature information, and then determine duplicate videos through feature comparison.

[0003] However, the inventors have found that when the above method is used to perform video duplicate detection, the following technical problems often exist:

[0004] First, the video duplicate detection is performed by comparing the feature similarity of each frame, which has large computing resource consumption and low duplicate detection efficiency and long duplicate detection time.

[0005] Second, in order to improve the representation ability of the extracted video frame feature information, it is often necessary to deepen the network depth of the feature extraction model, which leads to an increase in computing resource consumption and a longer feature extraction time.

[0006] Third, the representation ability of the video frame features extracted by the conventional convolution operation is poor, and the missing detection rate of material videos that have been processed by editing (such as view transformation and image cropping) is high, resulting in a large actual error in video duplicate detection.

[0007] The above information disclosed in the background section of this document is only intended to enhance the understanding of the background of the present inventive concept, and therefore, it can include information that does not form the prior art known to those of ordinary skill in the art in the country. SUMMARY

[0008] The summary section of the present disclosure is used to introduce the concept in a brief form, which will be described in detail in the specific embodiments section. The summary section of the present disclosure is not intended to identify key or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0009] Some embodiments of the present disclosure propose a material video duplicate detection method, device, electronic equipment and computer readable medium to solve one or more of the technical problems mentioned in the background section.

[0010] In a first aspect, some embodiments of the present disclosure provide a method for duplicate checking of a material video, the method comprising: receiving a material video uploaded by a user terminal as a video to be checked, wherein the video to be checked corresponds to video identification information and at least one video subject keyword label; selecting at least one material video from a material video library as a comparison material video according to the video to be checked and the at least one video subject keyword label, wherein the material video library is a database storing material videos uploaded by user terminals; analyzing the video to be checked to obtain a video frame sequence, wherein each video frame in the video frame sequence corresponds to video frame time information; for each comparison material video in the at least one comparison material video, performing the following duplicate checking screening steps: generating each key frame feature information of the video to be checked according to the comparison material video and the video frame sequence; determining the comparison material video as a similar material video in response to determining that the comparison material video satisfies a key frame matching condition according to each key frame feature information of the video to be checked; performing duplicate checking processing on the similar material video based on each target key frame of the video to be checked to determine material video duplicate checking information, wherein each target key frame is a video frame corresponding to each key frame feature information of the video to be checked that satisfies a key frame feature similarity condition; and sending each material duplicate checking information generated to the user terminal for displaying the material duplicate checking information by the user terminal.

[0011] In a second aspect, some embodiments of the present disclosure provide a material video duplication checking method and device. The device comprises: a receiving unit configured to receive a material video uploaded by a user terminal as a video to be checked for duplication, wherein the video to be checked for duplication corresponds to video identification information and at least one video subject keyword label; a selecting unit configured to select at least one material video from a material video library as each comparison material video according to the video to be checked for duplication and the at least one video subject keyword label, wherein the material video library is a database storing each material video uploaded by each user terminal; an analyzing unit configured to analyze the video to be checked for duplication to obtain a video frame sequence to be checked for duplication, wherein each video frame to be checked for duplication in the video frame sequence to be checked for duplication corresponds to video frame time information to be checked for duplication; a duplication checking and screening unit configured to, for each comparison material video in the each comparison material video, perform the following duplication checking and screening steps: generating each key frame feature information to be checked for duplication according to the comparison material video and the video frame sequence to be checked for duplication; determining the comparison material video as a similar material video in response to determining that the comparison material video meets a key frame matching condition according to the each key frame feature information to be checked for duplication; performing duplication processing on the similar material video based on each target key frame to be checked for duplication to determine material video duplication information, wherein the each target key frame to be checked for duplication is each video frame to be checked for duplication corresponding to each key frame feature information to be checked for duplication that meets a key frame feature similarity condition in the each key frame feature information to be checked for duplication; and a sending unit configured to send each material duplication information generated to the user terminal for displaying the each material duplication information by the user terminal.

[0012] In a third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the first aspect.

[0013] In a fourth aspect, some embodiments of the present disclosure provide a computer readable medium having a computer program stored thereon, wherein the program is executed by a processor to implement the method described in any implementation manner of the first aspect.

[0014] The above various embodiments of the present disclosure have the following beneficial effects: the material video duplicate checking method of some embodiments of the present disclosure can reduce the resource consumption of computing power and reduce the time consumption of video duplicate checking. Specifically, the reason for the large resource consumption of computing power and the long time consumption of video duplicate checking is that the video duplicate checking is performed by comparing the feature similarity of each frame, which consumes a large amount of computing power and has low duplicate checking efficiency and long duplicate checking time. Based on this, the material video duplicate checking method of some embodiments of the present disclosure first receives a material video uploaded by a user terminal as a duplicate checking video. The duplicate checking video corresponds to video identification information and at least one video subject keyword label. Then, at least one material video is selected from a material video library as each comparison material video according to the duplicate checking video and the at least one video subject keyword label. The material video library is a database storing each material video uploaded by each user terminal. Thus, each video that may be similar to or contain duplicate segments of the duplicate checking video can be initially selected as a comparison material video. Then, the duplicate checking video is analyzed to obtain a duplicate checking frame sequence. Each duplicate checking frame in the duplicate checking frame sequence corresponds to duplicate checking frame time information. Thus, the duplicate checking video can be analyzed to process images in the subsequent process as a basic unit. Next, for each comparison material video in the comparison material video, the following duplicate checking and screening steps are performed: first, each duplicate checking key frame feature information is generated according to the comparison material video and the duplicate checking frame sequence. Thus, by uniformly extracting key frames from the duplicate checking frame sequence and extracting features of each duplicate checking key frame, each duplicate checking key frame feature information representing the image content of each video frame can be obtained. Second, according to the duplicate checking key frame feature information, the comparison material video is determined as a similar material video in response to determining that the comparison material video satisfies the key frame matching condition. Thus, by comparing features a few times, it can be determined whether the comparison material video has a video frame similar to the duplicate checking key frame, so that the comparison material video can be preliminarily screened. Third, based on each target duplicate checking key frame, the similar material video is processed to determine material video duplicate checking information. The target duplicate checking key frame is a duplicate checking frame corresponding to each duplicate checking key frame feature information that satisfies the key frame feature similarity condition. Thus, when the comparison material video has a video frame similar to the duplicate checking key frame, the comparison material video is checked based on the duplicate checking key frame to generate material video duplicate checking information. Finally, each generated material duplicate checking information is sent to the user terminal for displaying the material duplicate checking information.In addition, the key frame extraction and key frame feature comparison are uniform, which can reduce the large amount of computing resources consumed by frame-by-frame feature extraction and frame-by-frame feature comparison of each comparison video and each video to be checked, thereby reducing the waste of computing resources and the time consumption of video checking. BRIEF DESCRIPTION OF DRAWINGS

[0015] The above and other features, aspects, and advantages of the present disclosure will become more apparent with reference to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, similar or same reference numerals are used to denote similar or same elements. It should be understood that the drawings are schematic and elements and features are not necessarily to scale.

[0016] Figure 1 is a flowchart of some embodiments of the material video checking method according to the present disclosure;

[0017] Figure 2 is a structural schematic diagram of some embodiments of the material video checking device according to the present disclosure;

[0018] Figure 3 is a structural schematic diagram of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0019] Embodiments of the present disclosure will be described in more detail with reference to the drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms, and should not be interpreted as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes, and are not intended to limit the scope of protection of the present disclosure.

[0020] In addition, it should be noted that only the parts related to the present invention are shown in the drawings for ease of description. The embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0021] It should be noted that the terms "first", "second", etc. mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0022] It should be noted that the adjectives "one", "multiple" mentioned in the present disclosure are illustrative and not limiting, and those skilled in the art should understand that unless otherwise explicitly stated in the context, it should be understood as "one or more".

[0023] Names of messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0024] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0025] Figure 1 Flow 100 of some embodiments of a material video duplicate detection method according to the present disclosure is shown. The material video duplicate detection method includes the following steps:

[0026] Step 101, receiving a material video uploaded by a user terminal as a video to be detected.

[0027] In some embodiments, the execution subject (for example, a computing device) of the material video duplicate detection method can receive a material video uploaded by a user terminal as a video to be detected through a wired connection or a wireless connection. Wherein, the video to be detected corresponds to video identification information and at least one video subject keyword label. The material video can be a video used as material for video editing and video creation. The video identification information can be a string generated by a digital signature algorithm representing the video to be detected. The at least one video subject keyword label can be each keyword selected by the user of the user terminal when uploading the video, representing the video subject of the video to be detected. For example, the at least one video subject keyword label can be "animation", "hot-blooded youth", and "inspiring".

[0028] It should be noted that the wireless connection can include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other now known or future developed wireless connection.

[0029] Step 102, selecting at least one material video from a material video library as each comparison material video according to the video to be detected and the at least one video subject keyword label.

[0030] In some embodiments, the execution subject can select at least one material video from a material video library as each comparison material video according to the video to be detected and the at least one video subject keyword label. Wherein, the material video library is a database storing each material video uploaded by each user terminal.

[0031] In some optional implementations of some embodiments, the execution subject can select at least one material video from a material video library as each comparison material video according to the video to be detected and the at least one video subject keyword label by the following steps:

[0032] In a first step, a video size of the video to be checked is determined as a video size of the video to be checked.

[0033] In a second step, a video duration of the video to be checked is determined as a video duration of the video to be checked. The video duration of the video to be checked has a time unit by default. The time unit can be a second.

[0034] In a third step, each material video corresponding to each video keyword tag in the at least one video keyword tag is selected from each material video included in the material video library as each comparison material video. The comparison material video can be a material video having at least one same video keyword tag as each video keyword tag corresponding to the video to be checked.

[0035] In a fourth step, in response to determining that there is at least one material video satisfying a size matching condition corresponding to the video size of the video to be checked in the material video library, the at least one material video satisfying the size matching condition is determined as each comparison material video. The size matching condition can be that the video size of the comparison material video is the same as the video size of the video to be checked.

[0036] In a fifth step, in response to determining that there is at least one material video satisfying a duration matching condition corresponding to the target video duration in the material video library, the at least one material video satisfying the duration matching condition is determined as each comparison material video. The duration matching condition can be that the video duration of the comparison material video is the same as the video duration of the video to be checked.

[0037] In step 103, the video to be checked is parsed to obtain a sequence of video frames to be checked.

[0038] In some embodiments, the execution subject can parse the video to be checked to obtain a sequence of video frames to be checked. Each video frame to be checked in the sequence of video frames to be checked corresponds to video frame time information to be checked. The video frame time information to be checked can be the time corresponding to each video frame to be checked in the video to be checked. In practice, the execution subject can split the video to be checked frame by frame by a video decoder to obtain the sequence of video frames to be checked. As an example, the video decoder can be an FFmpeg video decoder.

[0039] In step 104, for each comparison material video in each comparison material video, the following checking screening steps are performed:

[0040] In step 1041, each video frame to be checked key frame feature information is generated according to the comparison material video and the sequence of video frames to be checked.

[0041] In some embodiments, the execution subject can generate each to-be-duplicated key frame feature information according to the comparison material video and the to-be-duplicated frame sequence. The generated to-be-duplicated key frame feature information can be a video frame feature vector corresponding to the to-be-duplicated key frame. The to-be-duplicated key frame can be a key frame in the to-be-duplicated video sequence.

[0042] In some optional implementations of some embodiments, the execution subject can generate each to-be-duplicated key frame feature information according to the comparison material video and the to-be-duplicated frame sequence by the following steps:

[0043] Firstly, the video duration of the comparison material video is determined as a comparison video duration.

[0044] Secondly, a ratio of the comparison video duration to the duplication video duration is determined as a frame extraction adjustment coefficient.

[0045] Thirdly, according to the frame extraction adjustment coefficient and a frame extraction number corresponding to the duplication video duration, each to-be-duplicated video frame is selected from the to-be-duplicated frame sequence as each to-be-duplicated key frame. The frame extraction number can be a number of key frames to be extracted from the to-be-duplicated frame sequence. The frame extraction number can be a product of a preset unit time frame extraction amount and the duplication video duration. As an example, the preset unit time frame extraction amount can be 1 frame / second.

[0046] Fourthly, feature extraction processing is performed on each to-be-duplicated key frame to generate each to-be-duplicated key frame feature information.

[0047] In some optional implementations of some embodiments, the execution subject can select each to-be-duplicated video frame from the to-be-duplicated frame sequence as each to-be-duplicated key frame according to the frame extraction adjustment coefficient and the frame extraction number corresponding to the duplication video duration by the following steps:

[0048] Firstly, a product of the frame extraction adjustment coefficient and the frame extraction number is determined as a target frame extraction number.

[0049] Secondly, for each pair of adjacent video frames in the video frame sequence, optical flow feature extraction is performed on the two video frames to generate video frame optical flow information. The video frame optical flow information can be optical flow vectors of the two video frames. The generated video frame optical flow information corresponds to optical flow time information. The optical flow time information can be video frame time information corresponding to the first video frame in the pair of video frames. In practice, the execution subject can perform optical flow feature extraction on the two video frames by using a related library function (for example, the cv2.calcOpticalFlowPyrLK() method in the OpenCV library) to generate video frame optical flow information. The video frame optical flow information can represent the displacement direction and displacement amount of each pixel in the two video frames. When the video frame optical flow information is larger, it can represent that the image content change in the two video frames is larger.

[0050] Thirdly, the generated video frame optical flow information is sorted to obtain a video frame optical flow information sequence. In practice, the execution subject can sort the generated video frame optical flow information according to the time sequence to obtain a video frame optical flow information sequence.

[0051] Fourthly, each video frame optical flow information in the video frame optical flow information sequence that satisfies a preset optical flow threshold condition is determined as a target video frame optical flow information sequence. In practice, firstly, for each video frame optical flow information in the video frame optical flow information, the execution subject can determine the sum of squares of each vector element in the video frame optical flow information (i.e., the optical flow vector) as the total optical flow change amount. The preset optical flow threshold condition can be that the total optical flow change amount corresponding to the target video frame optical flow information is greater than or equal to a preset total optical flow change amount threshold. Then, the execution subject can determine each video frame optical flow information that satisfies the preset optical flow threshold condition as a target video frame optical flow information, and determine a target video frame optical flow information sequence according to the time sequence. Here, the specific setting of the preset total optical flow change amount threshold is not limited.

[0052] In the fifth step, a preset video start time and each video frame time information corresponding to the target video frame optical flow information sequence are determined as a target time information sequence. The preset video start time can be "00:00:00". For example, the target video frame optical flow information sequence can be (a30-31, a90-91, …, a300-301, a360-361). The "a23-24" can be the video frame optical flow information corresponding to the 30th video frame and the 31st video frame in the video frame sequence. The target time information sequence corresponding to the target video frame optical flow information sequence can be (00:00:00, 00:00:01, 00:00:03, …, 00:00:10, 00:00:12).

[0053] In the sixth step, based on each target time information in the target time information sequence, the following key frame extraction steps are performed:

[0054] In the first sub-step, in response to determining that the target time information does not satisfy the bit sequence condition, a frame segment time proportion is determined according to the target time information and a previous target time information. The previous target time information is the target time information adjacent to the target time information in the target time information sequence. The bit sequence condition can be that the target time information is the first target time information in the target time information sequence. In practice, first, the execution subject can determine a time difference between the target time information and the previous target time information as a time difference value. Then, the execution subject can determine a product of the time difference value and the ratio of the video length as the frame segment time proportion.

[0055] In the second sub-step, a product of the frame segment time proportion and the target frame number is determined as a frame segment frame number.

[0056] In the third sub-step, according to the target time and the previous target time, each video frame in the video frame sequence within the time range represented by the target time and the previous target time is extracted as a video key frame. In practice, first, the execution subject can determine a video frame sub-sequence in the video frame sequence within the time range represented by the target time and the previous target time. Then, the execution subject can uniformly extract (i.e., the same interval between the extracted video frames) each video frame in the determined video frame sub-sequence as a video key frame.

[0057] In some optional implementations of some embodiments, the execution subject can perform feature extraction processing on each video key frame to generate video key frame feature information by the following steps:

[0058] In the first step, for each of the above-mentioned to-be-checked key frames, the following multi-feature extraction steps are performed:

[0059] In the first sub-step, the above-mentioned to-be-checked key frame is subjected to local feature point detection processing to obtain key frame local feature position information. The obtained key frame local feature position information can be the image position coordinates of the corresponding feature points in the above-mentioned to-be-checked key frame. In practice, the above-mentioned execution subject can perform feature point detection on the above-mentioned to-be-checked key frame by using a pre-set feature point detection operator to obtain the key frame local feature position information. As an example, the above-mentioned feature point detection operator can be, but is not limited to, a DoG (Difference of Gaussians) operator.

[0060] In the second sub-step, based on the above-mentioned key frame local feature position information, the above-mentioned to-be-checked key frame is subjected to region cropping processing to obtain a key frame region image. In practice, for each of the above-mentioned key frame local feature position information, the above-mentioned execution subject can perform image cropping with the above-mentioned key frame local feature position information as the center and according to a pre-set cropping size to obtain a key frame region image. It should be noted that the above-mentioned pre-set cropping size can be a pre-set size, which is not limited herein.

[0061] In a third sub-step, the depth feature extraction processing is performed on each of the key frame region images to obtain key frame region feature information. The feature dimensions of the obtained key frame region feature information are equal. The key frame region feature information can be an image feature vector of the key frame region image. In practice, the execution subject can perform scaling processing on each of the key frame region images to uniformly scale to a preset input size to obtain a scaled key frame region image. As an example, the preset input size can be 32x32. Then, the execution subject can perform image preprocessing on each of the scaled key frame region images to obtain a processed key frame region image. The image preprocessing can include image normalization and grayscale. Finally, the execution subject can perform depth feature extraction processing on each of the processed key frame region images by using a pre-trained depth feature extraction model to obtain the key frame region feature information. The depth feature extraction model can be a neural network model that takes the processed key frame region image as input and outputs the corresponding key frame region feature information. The depth feature extraction model can include a first convolutional network, a second convolutional network, a third convolutional network, a fourth convolutional network, a fifth convolutional network, a sixth convolutional network, a seventh convolutional network, and an output layer. The first convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 32, and a step size of 1, a normalization layer, and a ReLu activation layer. The second convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 64, and a step size of 1 / 2, a normalization layer, and a ReLu activation layer. The third convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 64, and a step size of 1, a normalization layer, and a ReLu activation layer. The fourth convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 128, and a step size of 1 / 2, a normalization layer, and a ReLu activation layer. The fifth convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 128, and a step size of 1, a normalization layer, and a ReLu activation layer. The sixth convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 256, and a step size of 1 / 2, a normalization layer, and a ReLu activation layer. The seventh convolutional network can include a convolutional layer with a kernel size of 3x3, a channel number of 512, and a step size of 1 / 2, a normalization layer, and a ReLu activation layer. The output layer can include a convolutional layer with a kernel size of 8x8 and a step size of 1, a normalization layer, and a Droupout layer.

[0062] In a fourth sub-step, a key frame global feature matrix is determined based on the key frame region feature information. In practice, the execution subject can determine the matrix obtained by horizontally splicing the key frame region feature information as the key frame global feature matrix.

[0063] In a fifth sub-step, the key frame global feature matrix is dimensionally reduced to generate key frame feature information to be checked for duplication. In practice, the execution subject can perform dimensional reduction on the key frame global feature matrix by using a dimensional reduction algorithm to generate key frame feature information to be checked for duplication in a preset dimension. As an example, the dimensional reduction algorithm can be, but is not limited to, a PCA dimensional reduction algorithm. In this regard, the specific setting of the preset dimension is not limited.

[0064] The related content of the above embodiments can be used as one of the invention points of the embodiments of the present disclosure to solve the technical problem two, i.e., to improve the representation ability of the extracted video frame feature information, the network depth of the feature extraction model often needs to be deepened, thereby increasing the resource consumption of the algorithm and prolonging the feature extraction time. The factors that lead to the increase of the resource consumption of the algorithm and the prolongation of the feature extraction time are often as follows: to improve the representation ability of the extracted video frame feature information, the network depth of the feature extraction model often needs to be deepened. If the above factors are solved, the effect of reducing the resource consumption of the algorithm and reducing the feature extraction time can be achieved. In order to achieve this effect, the present disclosure generates video frame feature information by extracting each local feature in the video frame and performing feature fusion and feature dimension reduction on the local feature, thereby improving the representation ability of the generated video frame feature information, and further reducing the resource consumption of the algorithm and the feature extraction time.

[0065] In step 1042, according to each key frame feature information to be checked for duplication, the comparison material video is determined as a similar material video in response to determining that the comparison material video satisfies the key frame matching condition.

[0066] In some embodiments, the execution subject can determine the comparison material video as a similar material video in response to determining that the comparison material video satisfies the key frame matching condition according to each key frame feature information to be checked for duplication. The key frame matching condition can be that there is at least one comparison material video frame feature information satisfying the key frame feature similarity condition in the sequence of comparison material video frame feature information corresponding to the sequence of comparison material video frames.

[0067] In some optional implementations of some embodiments, the execution subject can determine the comparison material video as a similar material video in response to determining that the comparison material video satisfies the key frame feature similarity condition according to each key frame feature information to be checked for duplication by the following steps:

[0068] In a first step, the comparison material video is parsed to obtain a comparison material video frame sequence. Each comparison material video frame in the comparison material video frame sequence corresponds to comparison material video frame time information. The comparison material video frame time information can be the time of the corresponding comparison material video frame in the comparison material video. In practice, the execution subject can parse the comparison material video by using the video decoder to obtain the comparison material video frame sequence.

[0069] In a second step, feature extraction processing is performed on each comparison material video frame in the comparison material video frame sequence to generate each comparison material video frame feature information, thereby obtaining a comparison material video frame feature information sequence. It should be noted that the implementation of the feature extraction processing on each comparison material video frame in the comparison material video frame sequence to generate each comparison material video frame feature information can refer to the implementation of the generation of the to-be-checked key frame feature information, which will not be described here.

[0070] In a third step, for each to-be-checked key frame feature information in the to-be-checked key frame feature information, the following steps are performed:

[0071] In a first sub-step, the video frame feature similarity between the to-be-checked key frame feature information and each comparison material video frame feature information in the comparison material video frame feature information sequence is determined to obtain a video frame similarity set. In practice, the execution subject can determine the video frame feature similarity between the to-be-checked key frame feature information and each comparison material video frame feature information in the comparison material video frame feature information sequence by using a similarity algorithm to obtain a video frame similarity set. As an example, the similarity algorithm can be, but is not limited to, a cosine similarity algorithm.

[0072] In a first sub-step, in response to determining that there is at least one video frame similarity in the video frame similarity set that satisfies the key frame feature similarity condition, the following steps are performed:

[0073] Sub-step one, the at least one video frame similarity is determined as each target video frame similarity. The key frame feature similarity condition can be that the target video frame similarity is greater than or equal to a preset similarity threshold. As an example, the preset similarity threshold can be 0.9.

[0074] Sub-step two, for each target video frame similarity in the target video frame similarity, the to-be-checked video frame corresponding to the target video frame similarity is determined as a target to-be-checked key frame.

[0075] In a fourth step, the comparison material video is determined as a similar material video.

[0076] At step 1043, the similar material videos are checked for duplication based on each target key frame to be checked for duplication, to determine material video duplication information.

[0077] In some embodiments, the execution subject can check the similar material videos for duplication based on each target key frame to be checked for duplication, to determine material video duplication information. Each target key frame to be checked for duplication is a video frame corresponding to each key frame feature information to be checked for duplication that satisfies the key frame feature similarity condition in the key frame feature information to be checked for duplication.

[0078] In some optional implementations of some embodiments, the execution subject can check the similar material videos for duplication based on each target key frame to be checked for duplication, to determine material video duplication information, by the following steps.

[0079] First, for each target key frame to be checked for duplication in the target key frames to be checked for duplication, the following steps are performed.

[0080] First sub-step, according to the video frame time information corresponding to the target key frame to be checked for duplication, two target time information corresponding to the frame segment interception information are selected from the target time information sequence. In practice, the execution subject can select two target time information adjacent to the video frame time information corresponding to the target key frame to be checked for duplication from the target time information sequence as the frame segment interception information. For example, the video frame time information corresponding to the target key frame to be checked for duplication can be “00:00:21”. The target time sequence can be (00:00:00, …, 00:00:10, 00:00:40, …). The frame segment interception information corresponding to the video frame time information corresponding to the target key frame to be checked for duplication can be (00:00:10, 00:00:40).

[0081] Second sub-step, each video frame corresponding to the frame segment interception information is intercepted from the video frame sequence to be checked for duplication as a video frame sub-sequence to be checked for duplication. In practice, the execution subject can intercept each video frame within the time range represented by the frame segment interception information from the video frame sequence to be checked for duplication as a video frame sub-sequence to be checked for duplication.

[0082] A third sub-step, according to the comparison material video frames corresponding to the target video frame to be checked, each comparison material video frame is intercepted from the sequence of comparison material video frames as a comparison material video frame sub-sequence. Among them, the number of video frames to be checked in the video frame sub-sequence to be checked is the same as the number of comparison material video frames in the comparison material video frame sub-sequence. In practice, the execution subject can take the comparison material video frame corresponding to the target video frame to be checked as the center frame (i.e. the comparison material video frame that meets the key frame characteristic similarity condition), and intercept each comparison material video frame with the same total frame number (the number of comparison material video frames before and after the center frame is the same as the number of video frames before and after the target video frame to be checked) from the sequence of comparison material video frames as the comparison material video frame sub-sequence as the video frame sub-sequence to be checked.

[0083] A fourth sub-step, for each video frame to be checked in the video frame sub-sequence to be checked, the region feature extraction processing is performed on the video frame to be checked to generate video frame region feature information of each video frame to be checked. It should be noted that the implementation of the "region feature extraction processing is performed on the video frame to be checked to generate video frame region feature information of each video frame to be checked" can refer to the implementation of the "key frame region feature information", which will not be repeated again.

[0084] A fifth sub-step, according to the generated video frame region feature information of each video frame to be checked, a video frame region feature matching model is constructed. Among them, the video frame region feature matching model can be a tree model used for feature comparison. As an example, the video frame region feature matching model can be a KD tree model.

[0085] A sixth sub-step, according to the comparison material video frame sub-sequence, a material video region feature matching model is constructed. Among them, the material video region feature matching model can be a tree model constructed by each comparison material video frame region feature. As an example, the material video region feature matching model can be a KD tree model.

[0086] A seventh sub-step, according to the constructed material video region feature matching model and the constructed video frame region feature matching model, the frame segment similarity is determined. In practice, the execution subject can determine the structural similarity of the video frame region feature matching model and the material video region feature matching model as the frame segment similarity through a tree similarity algorithm. As an example, the tree similarity algorithm can be but not limited to the maximum common subtree algorithm, the topological distance algorithm and the tree edit distance algorithm.

[0087] An eighth sub-step, in response to determining that the frame segment similarity is greater than or equal to a preset similarity threshold, the comparison material video frame sub-sequence is determined as a duplicate comparison material video frame sub-sequence. As an example, the preset similarity threshold can be 0.80.

[0088] A ninth sub-step of determining the material video duplication information according to the determined duplicated comparison material video frame sub-sequence. In practice, the above-mentioned execution subject can determine the above-mentioned frame segment similarity and the above-mentioned frame segment clipping information as the material video duplication information.

[0089] The related content of the above-mentioned embodiment can solve the technical problem two “the representation ability of the video frame features extracted by the conventional convolution operation is poor, the missing rate of the material video after the editing processing (such as the perspective transformation, the image clipping) is high, and thus the actual error of the video duplication is large”. The factors leading to the actual error of the video duplication are often as follows: the representation ability of the video frame features extracted by the conventional convolution operation is poor, the missing rate of the material video after the editing processing (such as the perspective transformation, the image clipping) is high. If the above-mentioned factors are solved, the effect of reducing the video duplication error can be achieved. In order to achieve this effect, the video duplication is performed by the way of local feature extraction of the video frame sub-sequence and construction of the feature matching model using the extracted local features of each video frame. Thus, even if part of the video segments in the comparison material video are subjected to editing processing such as perspective transformation and image clipping, when each comparison material video frame and the corresponding video frame to be duplicated in the video segment contain highly similar local features, the video duplication can be determined, and thus the actual error of the video duplication can be greatly reduced.

[0090] Step 105, sending the generated each material duplication information to the user end for displaying each material duplication information by the user end.

[0091] In some embodiments, the above-mentioned execution subject can send the generated each material duplication information to the above-mentioned user end for displaying the above-mentioned each material duplication information by the above-mentioned user end. In practice, the above-mentioned execution subject can send the generated each material duplication information to the above-mentioned user end for displaying the above-mentioned each material duplication information on the associated display screen by the above-mentioned user end.

[0092] The above various embodiments of the present disclosure have the following beneficial effects: the material video duplicate checking method of some embodiments of the present disclosure can reduce the resource consumption of computing power and reduce the time consumption of video duplicate checking. Specifically, the reason for the large resource consumption of computing power and the long time consumption of video duplicate checking is that the video duplicate checking is performed by comparing the feature similarity of each frame, which consumes a large amount of computing power and has low duplicate checking efficiency and long duplicate checking time. Based on this, the material video duplicate checking method of some embodiments of the present disclosure first receives a material video uploaded by a user terminal as a duplicate checking video. The duplicate checking video corresponds to video identification information and at least one video subject keyword label. Then, at least one material video is selected from a material video library as each comparison material video according to the duplicate checking video and the at least one video subject keyword label. The material video library is a database storing each material video uploaded by each user terminal. Thus, each video that may be similar to or contain duplicate segments of the duplicate checking video can be initially selected as a comparison material video. Then, the duplicate checking video is analyzed to obtain a duplicate checking frame sequence. Each duplicate checking frame in the duplicate checking frame sequence corresponds to duplicate checking frame time information. Thus, the duplicate checking video can be analyzed to process images in the subsequent process as a basic unit. Next, for each comparison material video in the comparison material video, the following duplicate checking and screening steps are performed: first, each duplicate checking key frame feature information is generated according to the comparison material video and the duplicate checking frame sequence. Thus, by uniformly extracting key frames from the duplicate checking frame sequence and extracting features of each duplicate checking key frame, each duplicate checking key frame feature information representing the image content of each video frame can be obtained. Second, according to the duplicate checking key frame feature information, the comparison material video is determined as a similar material video in response to determining that the comparison material video satisfies the key frame matching condition. Thus, by comparing features a few times, it can be determined whether the comparison material video has a video frame similar to the duplicate checking key frame, so that the comparison material video can be preliminarily screened. Third, based on each target duplicate checking key frame, the similar material video is processed to determine material video duplicate checking information. The target duplicate checking key frame is a duplicate checking frame corresponding to each duplicate checking key frame feature information that satisfies the key frame feature similarity condition. Thus, when the comparison material video has a video frame similar to the duplicate checking key frame, the comparison material video is checked based on the duplicate checking key frame to generate material video duplicate checking information. Finally, each generated material duplicate checking information is sent to the user terminal for displaying the material duplicate checking information.Also because the uniform extraction of key frames and the comparison of key frame features are adopted, the large amount of computing resource consumption caused by the frame-by-frame feature extraction and the frame-by-frame feature comparison of the to-be-checked video and each comparison material video can be reduced, thereby reducing the waste of computing resource and the time consumption of video checking.

[0093] Further referring to Figure 2 , as an implementation of the method shown in the above figures, the disclosure provides some embodiments of a material video checking device, which correspond to the method embodiments shown in Figure 1 , and the material video checking device can be applied in various electronic devices.

[0094] As shown in Figure 2 , the material video checking device 200 of some embodiments includes a receiving unit 201, a selection unit 202, an analysis unit 203, a checking and screening unit 204, and a sending unit 205. The receiving unit 201 is configured to receive a material video uploaded by a user terminal as a to-be-checked video, wherein the to-be-checked video corresponds to video identification information and at least one video subject keyword label; the selection unit 202 is configured to select at least one material video from a material video library as each comparison material video according to the to-be-checked video and the at least one video subject keyword label, wherein the material video library is a database storing each material video uploaded by each user terminal; the analysis unit 203 is configured to analyze the to-be-checked video to obtain a to-be-checked video frame sequence, wherein each to-be-checked video frame in the to-be-checked video frame sequence corresponds to to-be-checked video frame time information; the checking and screening unit 204 is configured to, for each comparison material video in the comparison material videos, perform the following checking and screening steps: generating each to-be-checked key frame feature information according to the comparison material video and the to-be-checked video frame sequence; determining the comparison material video as a similar material video in response to determining that the comparison material video satisfies a key frame matching condition according to the each to-be-checked key frame feature information; performing checking and screening processing on the similar material video based on each target to-be-checked key frame to determine material video checking information, wherein the each target to-be-checked key frame is each to-be-checked video frame corresponding to each to-be-checked key frame feature information satisfying a key frame feature similarity condition in the each to-be-checked key frame feature information; and the sending unit 205 is configured to send each generated material checking information to the user terminal for displaying the each material checking information by the user terminal.

[0095] It can be understood that the units recorded in the material video checking device 200 correspond to the units recorded in the method embodiments shown in Figure 1The various steps in the described methods are correspondingly related. As such, the operations, features and benefits described above in relation to the methods are likewise applicable to the web page generation apparatus 200 and the units therein, in which they are similarly implemented.

[0096] Reference is made below to Figure 3 which shows a structural schematic diagram of an electronic device 300 suitable for use in implementing some embodiments of the present disclosure. Figure 3 The illustrated electronic device is merely one example, and should not be construed as limiting the scope of functionality or use of embodiments of the present disclosure.

[0097] As Figure 3 shown, the electronic device 300 can include a processing apparatus (e.g., a central processing unit, a graphics processing unit, etc.) 301 that can perform various appropriate actions and processes according to programs stored in a read-only memory 302 or loaded from a storage apparatus 308 into a random access memory 303. Various programs and data required for operation of the electronic device 300 are also stored in the random access memory 303. The processing apparatus 301, the read-only memory 302, and the random access memory 303 are connected to each other through a bus 304. An input / output interface 305 is also connected to the bus 304.

[0098] Generally, the following apparatuses can be connected to the I / O interface 305: input apparatuses 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output apparatuses 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage apparatuses 308 including, for example, a magnetic tape, a hard disk, etc.; and communication apparatuses 309. The communication apparatuses 309 can allow the electronic device 300 to perform wireless or wired communication with other devices to exchange data. Although Figure 3 The electronic device 300 is shown with various apparatuses, but it should be understood that not all of the shown apparatuses are required to be implemented or present. More or fewer apparatuses can alternatively be implemented or present. Figure 3 Each block shown in the flowcharts can represent an apparatus or, as desired, multiple apparatuses.

[0099] In particular, processes described above with reference to the flowcharts can be implemented as a computer software program according to some embodiments of the present disclosure. For example, some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In some such embodiments, the computer program can be downloaded and installed from a network through the communication apparatuses 309, or installed from the storage apparatuses 308, or installed from the read-only memory 302. When the computer program is executed by the processing apparatus 301, the above-mentioned functions defined in the methods of some embodiments of the present disclosure are performed.

[0100] Note that the computer readable medium in some embodiments of the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination thereof. The computer readable storage medium may, for example and without limitation, be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the present disclosure, the computer readable storage medium can be any tangible medium that contains or stores a program used by or in connection with an instruction execution system, apparatus or device. In some embodiments of the present disclosure, the computer readable signal medium can include a computer readable program code propagated in or on a carrier medium, in which the computer readable program code is embodied. Such propagated computer readable program code can take many forms, including but not limited to electro-magnetic signals, optical signals, or any suitable combination thereof. The computer readable signal medium can also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport a program for use by or in connection with an instruction execution system, apparatus or device. Program code embodied on a computer readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0101] In some embodiments, the client, server, or both can communicate using any current known or future developed network protocol, such as HTTP (Hyper Text Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet, and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any current known or future developed networks.

[0102] The computer readable medium can be included in the electronic device; or can exist separately from the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: receive a material video uploaded by a user terminal as a video to be checked for duplication, wherein the video to be checked for duplication corresponds to video identification information and at least one video subject keyword label; select at least one material video from a material video library as each comparison material video according to the video to be checked for duplication and the at least one video subject keyword label, wherein the material video library is a database storing each material video uploaded by each user terminal; analyze the video to be checked for duplication to obtain a video to be checked for duplication frame sequence, wherein each video to be checked for duplication frame in the video to be checked for duplication frame sequence corresponds to video to be checked for duplication frame time information; for each comparison material video in the each comparison material video, perform the following duplication checking and screening steps: generate each video to be checked for duplication key frame feature information according to the comparison material video and the video to be checked for duplication frame sequence; determine the comparison material video as a similar material video in response to determining that the comparison material video satisfies a key frame matching condition according to the each video to be checked for duplication key frame feature information; perform duplication processing on the similar material video based on each target video to be checked for duplication key frame to determine material video duplication information, wherein the each target video to be checked for duplication key frame is each video to be checked for duplication frame corresponding to each video to be checked for duplication key frame feature information satisfying a key frame feature similarity condition in the each video to be checked for duplication key frame feature information; and send each generated material duplication information to the user terminal for the user terminal to display the each material duplication information.

[0103] Computer program code for carrying out operations of some embodiments of the disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0104] The computer program product of the first aspect can include a computer readable storage medium. The computer readable storage medium can include instructions. The instructions can include one or both of: instructions for causing a computer to implement a receiving unit, a selecting unit, a parsing unit, a duplicate screening unit, and a sending unit. The instructions can be executable by one or more processors of a computer. The one or more processors can be configured to implement the receiving unit, the selecting unit, the parsing unit, the duplicate screening unit, and the sending unit.

[0105] The units described in some embodiments of the present disclosure can be implemented in the form of software, or can be implemented in the form of hardware. The described units can also be arranged in a processor, for example, can be described as: a processor includes a receiving unit, a selecting unit, a parsing unit, a duplicate screening unit and a sending unit. Among them, the names of these units do not constitute a limitation to the units themselves in some cases, for example, the receiving unit can also be described as "a unit for receiving a material video uploaded by a user terminal as a video to be screened for duplicates".

[0106] The functions described above in the present document can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, example types of hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip systems (SOCs), complex programmable logic devices (CPLDs), etc.

[0107] The above description is merely some of the preferred embodiments of the present disclosure and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combinations of the above technical features or their equivalent features without departing from the above inventive concept. For example, the above features are replaced with the technical features disclosed in the embodiments of the present disclosure (but not limited to) having similar functions to form technical solutions.

Claims

1. A material video duplicate detection method, comprising: receiving a material video uploaded by a user terminal as a video to be detected, wherein the video to be detected corresponds to video identification information and at least one video subject keyword label; selecting at least one material video from a material video library as a respective comparison material video according to the video to be detected and the at least one video subject keyword label, wherein the material video library is a database storing respective material videos uploaded by respective user terminals; analyzing the video to be detected to obtain a video frame sequence to be detected, wherein each video frame to be detected in the video frame sequence to be detected corresponds to video frame time information to be detected; for each comparison material video in the respective comparison material video, performing the following duplicate detection screening steps: generating respective key frame feature information to be detected according to the comparison material video and the video frame sequence to be detected, comprising: determining a video duration of the comparison material video as a comparison video duration; determining a ratio of the comparison video duration and the duplicate video duration as a frame extraction adjustment coefficient; selecting respective video frames to be detected from the video frame sequence to be detected as respective key frames to be detected according to the frame extraction adjustment coefficient and a frame number corresponding to the duplicate video duration, comprising: determining a product of the frame extraction adjustment coefficient and the frame number as a target frame number; for each adjacent two video frames to be detected in the video frame sequence to be detected, performing optical flow feature extraction on the two video frames to be detected to generate video frame optical flow information to be detected, wherein the generated video frame optical flow information to be detected corresponds to optical flow time information; sorting the generated respective video frame optical flow information to be detected to obtain a video frame optical flow information sequence to be detected; determining respective video frame optical flow information to be detected in the video frame optical flow information sequence to be detected that satisfy a preset optical flow threshold condition as a target video frame optical flow information sequence; determining a preset video start time and respective optical flow time information corresponding to the target video frame optical flow information sequence as a target time information sequence; based on each target time information in the target time information sequence, performing the following key frame extraction steps: in response to determining that the target time information does not satisfy a bit sequence condition, determining a frame segment proportion of time according to the target time information and a previous target time information, wherein the previous target time information is a previous target time information adjacent to the target time information; determining a product of the frame segment proportion of time and the target frame number as a frame segment frame number; extracting respective video frames to be detected equal to the frame segment frame number from the video frame sequence to be detected as respective key frames to be detected according to the target time and the previous target time; performing feature extraction processing on the respective key frames to be detected to generate respective key frame feature information to be detected; in response to determining that the comparison material video satisfies a key frame matching condition according to the respective key frame feature information to be detected, determining the comparison material video as a similar material video. based on each target key frame to be checked for duplication, performing a duplication checking process on the similar material video to determine material video duplication checking information, wherein the each target key frame to be checked for duplication is each video frame to be checked for duplication corresponding to each key frame feature information to be checked for duplication that meets a key frame feature similarity condition in the each key frame feature information to be checked for duplication; sending the each material duplication checking information generated to the user end for the user end to display the each material duplication checking information.

2. The method of claim 1, wherein, The selecting, from a material video library, at least one material video as each comparison material video according to the video to be checked for duplication and the at least one video subject keyword label includes: determining a video size of the video to be checked for duplication as a duplication checking video size; determining a video duration of the video to be checked for duplication as a duplication checking video duration; selecting, from each material video included in the material video library, each material video corresponding to the at least one video subject keyword label as each comparison material video; in response to determining that there is at least one material video in the material video library that meets a size matching condition corresponding to the duplication checking video size, determining the at least one material video that meets the size matching condition as each comparison material video; in response to determining that there is at least one material video in the material video library that meets a duration matching condition corresponding to the target video duration, determining the at least one material video that meets the duration matching condition as each comparison material video.

3. The method of claim 2, wherein, The determining, according to the each key frame feature information to be checked for duplication, the comparison material video as a similar material video in response to determining that the comparison material video meets a key frame matching condition includes: parsing the comparison material video to obtain a comparison material video frame sequence, wherein a comparison material video frame in the comparison material video frame sequence corresponds to comparison material video frame time information; performing feature extraction processing on each comparison material video frame in the comparison material video frame sequence to generate each comparison material video frame feature information, thereby obtaining a comparison material video frame feature information sequence; for each key frame feature information to be checked for duplication in the each key frame feature information to be checked for duplication, the following steps are performed: determining a video frame feature similarity between the key frame feature information to be checked for duplication and each comparison material video frame feature information in the comparison material video frame feature information sequence, thereby obtaining a video frame similarity set; in response to determining that there is at least one video frame similarity that meets a key frame feature similarity condition in the video frame similarity set, the following steps are performed: determining the at least one video frame similarity as each target video frame similarity; for each target video frame similarity in the each target video frame similarity, determining a video frame to be checked for duplication corresponding to the target video frame similarity as a target key frame to be checked for duplication; determining the comparison material video as a similar material video.

4. A material video duplication checking apparatus, comprising: a receiving unit configured to receive a material video uploaded by a user end as a video to be checked for duplication, wherein the video to be checked for duplication corresponds to video identification information and at least one video subject keyword label; The selecting unit is configured to select at least one material video as each contrast material video from a material video library according to the video to be checked and the at least one video subject keyword label, wherein the material video library is a database storing each material video uploaded by each user terminal; The analyzing unit is configured to analyze the video to be checked to obtain a video to be checked frame sequence, wherein each video to be checked frame in the video to be checked frame sequence corresponds to video to be checked frame time information; The duplicate screening unit is configured to perform the following duplicate screening steps for each of the respective comparison material videos: generating respective to-be-screened key frame feature information according to the comparison material video and the to-be-screened video frame sequence, including: determining the video duration of the comparison material video as a comparison video duration; determining the ratio of the comparison video duration and the duplicate video duration as a frame extraction adjustment coefficient; selecting respective to-be-screened video frames from the to-be-screened video frame sequence as respective to-be-screened key frames according to the frame extraction adjustment coefficient and the frame extraction quantity corresponding to the duplicate video duration, including: determining the product of the frame extraction adjustment coefficient and the frame extraction quantity as a target frame extraction quantity; for every two adjacent to-be-screened video frames in the to-be-screened video frame sequence, performing optical flow feature extraction on the two to-be-screened video frames to generate to-be-screened video frame optical flow information, wherein the generated to-be-screened video frame optical flow information corresponds to optical flow time information; sorting the generated respective to-be-screened video frame optical flow information to obtain a to-be-screened video frame optical flow information sequence; determining respective to-be-screened video frame optical flow information in the to-be-screened video frame optical flow information sequence that satisfy a preset optical flow threshold condition as a target video frame optical flow information sequence; determining a preset video start time and respective optical flow time information corresponding to the target video frame optical flow information sequence as a target time information sequence; based on each target time information in the target time information sequence, performing the following key frame extraction steps: in response to determining that the target time information does not satisfy a bit sequence condition, determining a frame segment occupancy ratio according to the target time information and a previous target time information; determining the product of the frame segment occupancy ratio and the target frame extraction quantity as a frame segment frame extraction quantity; extracting respective to-be-screened video frames equal to the frame segment frame extraction quantity from the to-be-screened video frame sequence as respective to-be-screened key frames according to the target time and the previous target time; performing feature extraction processing on the respective to-be-screened key frames to generate respective to-be-screened key frame feature information; in response to determining that the comparison material video satisfies a key frame matching condition according to the respective to-be-screened key frame feature information, determining the comparison material video as a similar material video; based on respective target to-be-screened key frames, performing duplicate screening processing on the similar material video to determine material video duplicate screening information, wherein the respective target to-be-screened key frames are respective to-be-screened video frames corresponding to respective to-be-screened key frame feature information that satisfy a key frame feature similarity condition in the respective to-be-screened key frame feature information; The sending unit is configured to send the generated respective material duplicate screening information to the user end for the user end to display the respective material duplicate screening information.

5. An electronic device, comprising: one or more processors; storage having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement a method as claimed in any of claims 1 to 3.

6. A computer readable medium having stored thereon a computer program, wherein, The computer program, which when executed by a processor, implements a method as claimed in any of claims 1 to 3.

Images