Video segment determination method, live body detection method, electronic device, and storage medium

Abstract

The present disclosure provides a video segment determination method, a live detection method, an electronic device and a storage medium, and is applied to the technical field of data processing. The method can extract a first audio signal sequence in a target video, wherein the target video is a video collected when a user reads target content, and the first audio signal sequence includes audio signal values corresponding to each sampling point; then determine a segmentation threshold according to the first audio signal sequence, wherein the segmentation threshold is a critical value of the audio signal corresponding to the case that the sampling point belongs to the video segment of reading the target content; and then determine the video segment corresponding to the user reading the target content according to the segmentation threshold and the first audio signal sequence. The method uses the characteristics of audio and video timing correspondence to analyze the audio signal to determine the video segment in the target video, which is simple to implement, has low requirements for video quality and hardware deployment, and is good in universality; the segmentation threshold is determined based on the distribution of the audio signal value, and the segmentation rule is uniform, the extraction cost is low, and the robustness is good.

Description

Technical Field

[0001] This disclosure belongs to the field of data processing technology, specifically relating to a method for determining video segments, a method for detecting liveness, an electronic device, and a storage medium. Background Technology

[0002] Liveness detection is a biometric method used to verify whether a subject is alive. It effectively defends against attacks using photos, videos, or occlusion techniques in facial recognition verification scenarios. Lip-reading liveness detection is a reading-based method that acquires a video of a user reading out randomly selected numbers, extracts lip movement features from the video, and performs liveness detection based on these lip movements.

[0003] However, videos typically contain multiple frames. Extracting and recognizing features from each frame is computationally intensive, inefficient, and costly. Furthermore, background frames can interfere with feature extraction, leading to low recognition accuracy. Therefore, accurately locating the reading segment within the entire video is crucial.

[0004] Currently, in the entire video, one can manually define video reading segments, but this method requires the introduction of hyperparameters, and it is difficult to unify segmentation rules for different recognition scenarios, resulting in high extraction costs and poor robustness. Alternatively, reading segments can be determined based on the video frame, but this method relies on image processing technology to analyze the user's lip movements, resulting in a complex algorithm structure, high implementation difficulty, and requires clear and smooth video. It also has high hardware deployment costs for recording, shooting, and transmission, and poor versatility. Summary of the Invention

[0005] The purpose of this disclosure is to provide a video segment determination method, a liveness detection method, an electronic device, and a storage medium that can solve the problems of high extraction cost and poor robustness when extracting video segments by manually introducing hyperparameters, or the problems of complex algorithm structure, high requirements for video quality, and poor versatility when relying on image processing technology for image analysis.

[0006] To solve the above-mentioned technical problems, this disclosure is implemented as follows:

[0007] In a first aspect, this disclosure provides a method for determining video segments, the method comprising: extracting a first audio signal sequence from a target video; wherein the target video is a video captured when a user reads target content, and the first audio signal sequence includes audio signal values ​​corresponding to each sampling point; determining a segmentation threshold based on the first audio signal sequence; wherein the segmentation threshold is a critical value of the audio signal corresponding to the sampling point when it belongs to a video segment of the target content being read;

[0008] Based on the segmentation threshold and the first audio signal sequence, determine the location information of the sampling point corresponding to when the user reads the target content; based on the location information, determine the video segment corresponding to the user reading the target content from the target video.

[0009] Optionally, determining the segmentation threshold based on the first audio signal sequence includes: filtering the first audio signal sequence to obtain a second audio signal sequence; sorting the audio signal values ​​in the second audio signal sequence to obtain a third audio signal sequence; and determining the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence.

[0010] Optionally, the location information includes a start position and an end position. Determining the location information of the sampling point corresponding to the user reading the target content based on the segmentation threshold and the first audio signal sequence includes: determining the position corresponding to the first sampling point as the start position along the temporal direction of the first audio signal sequence, where the first sampling point is the first sampling point whose audio signal value is greater than or equal to the segmentation threshold; and determining the position corresponding to the second sampling point as the end position, where the second sampling point is the last sampling point whose audio signal value is greater than or equal to the segmentation threshold.

[0011] Optionally, the step of filtering the first audio signal sequence to obtain the second audio signal sequence includes: averaging the audio signal values ​​in the first audio signal sequence to obtain a first average value; averaging the audio signal values ​​in the first audio signal sequence that are greater than the first average value to obtain a second average value; averaging the audio signal values ​​in the first audio signal sequence that are less than the first average value to obtain a third average value; assigning the audio signal values ​​in the first audio signal sequence that are greater than the second average value to the second average value, and assigning the audio signal values ​​that are less than the third average value to the third average value, thereby obtaining the second audio signal sequence.

[0012] Optionally, sorting the audio signal values ​​in the second audio signal sequence to obtain the third audio signal sequence includes: sorting the audio signal values ​​in the second audio signal sequence in descending order of value to obtain the third audio signal sequence; correspondingly, determining the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence includes: determining the difference corresponding to each audio signal value in the third audio signal sequence, the difference being obtained by subtracting each audio signal value from its adjacent audio signal values ​​along the sorting direction; when the difference meets the rate of change of the second average value relative to the third average value, the audio signal value corresponding to the difference is used as the segmentation threshold.

[0013] Optionally, determining the video segment corresponding to the user's target content from the target video based on the location information includes: defraction of the target video to obtain a video frame sequence, the video frame sequence corresponding to the first audio signal sequence in the temporal direction; determining a start frame in the temporal direction of the video frame sequence based on the start position; determining an end frame in the temporal direction of the video frame sequence based on the end position; and determining the video segment corresponding to the user's target content from the video frame sequence based on the start frame and the end frame.

[0014] Secondly, this disclosure provides a liveness detection method, which may include: acquiring a target video of a user reading target content; using the video segment determination method as described in the first aspect to determine the video segment corresponding to the user reading the target content from the target video; and performing liveness detection on the user based on the video segment.

[0015] Thirdly, this disclosure provides an electronic device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the video segment determination method of the first aspect.

[0016] Fourthly, this disclosure provides a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the video segment determination method of the first aspect.

[0017] Fifthly, this disclosure provides a chip including a processor and a communication interface coupled to the processor, the processor being used to run programs or instructions to implement the steps of the video segment determination method of the first aspect.

[0018] In a sixth aspect, this disclosure provides a computer program product that, when run on an electronic device, causes the electronic device to perform steps such as those implementing the video segment determination method of the first aspect.

[0019] In the video segment determination method provided in this disclosure, a first audio signal sequence can be extracted from the target video, wherein the target video is the video captured when the user reads the target content, and the first audio signal sequence includes the audio signal values ​​corresponding to each sampling point; further, a segmentation threshold is determined based on the first audio signal sequence, the segmentation threshold being the critical value of the audio signal corresponding to the sampling point belonging to the video segment of the target content being read; then, the video segment corresponding to the target content read by the user is determined based on the segmentation threshold and the first audio signal sequence. In this method, the video segments in the target video are determined based on the analysis of the audio signal, utilizing the temporal correspondence between audio and video in the target video. It does not rely on image processing technology, does not require complex algorithm structures, is simple to implement, has low requirements for video quality, and thus has low requirements for hardware deployment such as recording, shooting, and transmission, and has good versatility; at the same time, the segmentation threshold is determined based on the distribution of audio signal values ​​in the audio sequence, without the need for manual introduction or specification of hyperparameters, and parameter tuning is performed in different scenarios, resulting in unified segmentation rules, low extraction cost, and good robustness. Attached Figure Description

[0020] Figure 1 This is one of the flowcharts of the video segment determination method provided in the embodiments of this disclosure;

[0021] Figure 2 This is the second flowchart of the video segment determination method provided in the embodiments of this disclosure;

[0022] Figure 3 This is a schematic diagram of the normalization processing of the first audio signal sequence provided in an embodiment of the present disclosure;

[0023] Figure 4 A schematic diagram of the first audio signal sequence denoising process provided in an embodiment of this disclosure;

[0024] Figure 5 A schematic diagram of a second audio signal sequence provided in an embodiment of this disclosure;

[0025] Figure 6 A schematic diagram of a third audio signal sequence provided in an embodiment of this disclosure;

[0026] Figure 7 This is a schematic diagram illustrating the location information determination provided in an embodiment of the present disclosure;

[0027] Figure 8 A flowchart illustrating the steps of the liveness detection method provided in this embodiment of the disclosure;

[0028] Figure 9 A structural block diagram of the video segment determination device provided in the embodiments of this disclosure;

[0029] Figure 10 This is a structural block diagram of the liveness detection device provided in the embodiments of this disclosure;

[0030] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure;

[0031] Figure 12 This is a hardware schematic diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0032] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this disclosure. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0033] The terms "first," "second," etc., used in this disclosure and in the claims are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this disclosure can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0034] It should be noted that the data obtained in this public disclosure, including video, audio and other related data, are accessed, collected, stored and used for subsequent analysis and processing after the user or the relevant data owner has been clearly informed of the content of the data collection, the purpose of the data, the processing method and other information, and with the consent and authorization of the user or the relevant data owner. Furthermore, the public may provide the user or the relevant data owner with the means to access, correct or delete the data, as well as the method to revoke consent or authorization.

[0035] The determination of video segments provided in this disclosure will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0036] Figure 1 This is one of the flowcharts for a video segment determination method provided in an embodiment of the present disclosure. The method may include the following steps 101 to 105.

[0037] Step 101: Extract the first audio signal sequence from the target video; wherein, the target video is the video captured when the user reads the target content, and the first audio signal sequence includes the audio signal values ​​corresponding to each sampling point.

[0038] The target video can be a video captured when a user reads target content. For example, the target video could be a video of readings taken during a liveness detection process, or a video of the test subject reading random numbers. Alternatively, the target video could be a video of readings taken during a language test, or a video of the test subject reading test text. In this case, the video segment of the target video in which the user reads the target content can be used as the foreground, and the rest as the background. The signal intensity of the audio signal in the target video will change from background to foreground and from foreground to background, thus allowing the foreground video segment to be determined in the target video based on the analysis of the audio signal in this embodiment.

[0039] In this embodiment of the disclosure, extracting the first audio signal sequence from the target video can be achieved by sampling audio along the temporal direction of the target video to obtain audio signal values ​​corresponding to different sampling points. The number of samplings, frequency, etc., can be set according to actual needs, sampling conditions, target video length, etc. In optional embodiments of this disclosure, extracting the first audio signal sequence from the target video can be implemented using audio extraction tools, such as FFmpeg (Fast Forward Moving Picture Experts Group). The extracted first audio signal sequence can be in audio formats such as WAV (WaveForm), MP3 (Moving Picture Experts Group Audio Layer III), FLAC (Free Lossless Audio Codec), etc. The first audio signal sequence can be a digital sequence containing audio signal values, where the sequence length can represent the audio length and the value can represent the signal strength. Those skilled in the art can choose the method for extracting the first audio signal sequence and the format of the audio signal sequence according to data processing conditions, subsequent task requirements, etc. This embodiment of the disclosure does not impose specific limitations on this.

[0040] Step 102: Determine the segmentation threshold based on the first audio signal sequence; wherein, the segmentation threshold is the critical value of the audio signal corresponding to the sampling point belonging to the video segment of the target content.

[0041] In this embodiment, a segmentation threshold can be determined based on the audio signal value in the first audio signal sequence. The segmentation threshold is the audio signal threshold corresponding to the case where the sampling point belongs to the video segment containing the target content. Since the video segment containing the target content is the foreground portion of the target video with a high audio signal intensity, when the audio signal value corresponding to a sampling point is less than an audio signal threshold, the sampling point does not belong to the video segment containing the target content. This audio signal threshold can then be used as the segmentation threshold between the foreground and background portions of the target video. The segmentation threshold is determined based on the intensity change analysis of the audio signal, according to the change in audio signal values ​​between the foreground and background portions of the target video. This depends on the audio distribution characteristics of the target video itself, without introducing any additional hyperparameters.

[0042] Step 103: Determine the location information of the sampling point corresponding to the user reading the target content based on the segmentation threshold and the first audio signal sequence.

[0043] In this embodiment, since the audio signal value can represent the intensity of the audio signal at the sampling point location, and the audio signal sequence has a temporal correspondence with the target video, when the audio signal value in the first audio signal sequence meets the segmentation threshold, it can be considered that the corresponding sampling point belongs to the video segment when the user reads the target content. Furthermore, the position information of the sampling point corresponding to the user reading the target content in the first audio signal sequence can be determined. Based on this, the position information of the video segment corresponding to the user reading the target content in the target video can be determined based on the temporal correspondence characteristic.

[0044] Step 104: Determine the video segment corresponding to the target content read by the user from the target video based on the location information.

[0045] In this embodiment of the disclosure, since the audio signal sequence and video frames in the target video correspond in time, the video segment at the corresponding time position can be determined in the target video based on the position information of the sampling point corresponding to the user reading the target content in the first audio signal sequence. Thus, the video segment corresponding to the user reading the target content can be determined in the target video. Therefore, without relying on image processing technology or manually introducing hyperparameters, the video segment in the target video can be determined based on the analysis of the audio signal.

[0046] Figure 2 This is a second flowchart illustrating the steps of a video segment determination method provided in this embodiment of the disclosure. The method may include steps 201 to 209.

[0047] Step 201: Extract the first audio signal sequence from the target video; wherein the target video is the video collected when the user reads the target content, and the first audio signal sequence includes the audio signal values ​​corresponding to each sampling point.

[0048] In this embodiment of the disclosure, step 201 can be referred to the relevant description of step 101 above. To avoid repetition, it will not be repeated here.

[0049] Furthermore, to control the amount of data and avoid excessive computation that could affect processing efficiency and results, the required number of sampling points can be preset, and the first audio signal sequence can be extracted from the target video based on the number of sampling points. If the total number of audio signal sequences in the target video is less than the preset number of sampling points, the number of sampling points can be reset to the length of the total number of audio signal sequences, and the first audio signal sequence can be extracted from the target video based on this number of sampling points.

[0050] For example, if the number of sampling points N is preset to 100,000, then when the length of all audio signal sequences corresponding to the target video is less than or equal to 10,000, sampling is performed based on N to obtain the first audio signal sequence.

[0051] In one optional embodiment of this disclosure, the first audio signal sequence is typically a digital sequence of uncertain length and value, which may suffer from inconsistent format and scattered numerical distribution, affecting the efficiency and effectiveness of subsequent processing. Therefore, the first audio signal sequence can be preprocessed, including normalization and denoising. Normalization limits each value in the first audio signal sequence to a predetermined range, thereby reducing the impact of scattered values ​​on the processing results while preserving the overall information of the sequence. It also avoids computational processing of complex data, facilitating further analysis. Denoising effectively removes noise from the first audio signal sequence, allowing it to retain more feature information and preventing noise from interfering with the data analysis results.

[0052] In one optional embodiment of the method disclosed herein, the normalization process may include linear normalization.

[0053] In this embodiment, linear normalization (Min-Max Normalization) can be used to normalize the first audio signal sequence. Linear normalization, also known as deviation normalization or extremum normalization, involves first obtaining the maximum and minimum values, as well as the difference between the maximum and minimum values. Then, the minimum value is subtracted from each value, and the result is compared with this difference. This ratio is used as the normalized result for that value. Linear normalization scales the values ​​proportionally, reducing the burden of subsequent processing and computation, limiting the range of value distribution, minimizing the impact of extreme values ​​on the overall sequence properties, and improving analysis efficiency and accuracy.

[0054] Figure 3 This is a schematic diagram of the normalization processing of the first audio signal sequence provided in an embodiment of this disclosure, as shown below. Figure 3 As shown, the horizontal axis represents the number of sampling points, and the vertical axis represents the normalization result. The first audio signal sequence with a sequence length of N is linearly normalized so that its values ​​are distributed in the interval [0,1], and the overall mean is distributed around 0.5.

[0055] In one optional embodiment of the method disclosed herein, the denoising process may include at least one of wavelet transform denoising, median filtering, low-pass filtering, and high-pass filtering.

[0056] In this embodiment, wavelet transform denoising, median filtering, low-pass filtering, high-pass filtering, and other methods can be used to denoise the first audio signal sequence. Depending on the type of noise to be removed, the requirements for subsequent audio signal processing, and the computational conditions, those skilled in the art can choose the corresponding denoising method; this embodiment does not impose specific limitations in this regard. Wavelet transform denoising can perform localized analysis of time and spatial frequencies, progressively refining the audio signal at multiple scales through scaling and translation operations. This achieves time subdivision at high frequencies and frequency subdivision at low frequencies, automatically adapting to the analysis needs of time-frequency signals and flexibly focusing on detailed information in the audio signal.

[0057] Figure 4 This is a schematic diagram of the first audio signal sequence denoising process provided in an embodiment of this disclosure, as shown below. Figure 4 As shown, the horizontal axis represents the number of sampling points, and the vertical axis represents the normalization result. Figure 3 After the normalization process shown, it can be seen that there is still a lot of noise interference in the first audio signal sequence. At this time, wavelet transform can be used to further denoise the first audio signal, thereby obtaining the desired result. Figure 4 The first audio signal sequence shown is [x0,x1,…,x] n The sequence length N is 10000, the values ​​are distributed in the interval [0,1], and the overall mean is distributed around 0.5. The noise distribution is low and the audio signal is clear and effective.

[0058] Step 202: Filter the first audio signal sequence to obtain the second audio signal sequence.

[0059] In this embodiment, due to the varying energy concentration of different audio signals in the first audio signal sequence, the numerical distribution of the signals is relatively dispersed. This may result in larger and smaller peaks having a higher degree of dominance over the overall attributes, thus affecting the accuracy of audio signal analysis. Therefore, after extracting the first audio signal sequence from the target video, the first audio signal sequence can be filtered to remove some peaks that may interfere with the accuracy of the results, making the audio signal value distribution in the obtained second audio signal sequence more concentrated than that in the first audio signal sequence. Here, concentration refers to the degree of concentration of the values ​​in the sequence towards a central location, which can be a value reflecting the overall level of the values ​​in the sequence, such as the median, mean, or mode of the values ​​in the sequence. By filtering the first audio signal sequence, some larger and smaller peaks can be removed, making the concentration of the numerical distribution in the second audio signal sequence greater than that in the first audio signal sequence. The more concentrated distribution of numbers in the sequence can reduce interference in subsequent signal processing.

[0060] In an optional embodiment of the method disclosed herein, the filtering process may be a mean filtering process, then step 202 may include the following steps A1 to A3.

[0061] Step A1: Perform mean processing on the audio signal values ​​in the first audio signal sequence to obtain the first average value.

[0062] Step A2: Average the audio signal values ​​in the first audio signal sequence that are greater than the first average value to obtain the second average value, and average the audio signal values ​​in the first audio signal sequence that are less than the first average value to obtain the third average value.

[0063] Step A3: Assign the audio signal values ​​in the first audio signal sequence that are greater than the second average value to the second average value, and assign the audio signal values ​​that are less than the third average value to the third average value to obtain the second audio signal sequence.

[0064] In this embodiment, symmetrical mean filtering can be used. In the first audio signal sequence, the overall audio signal values ​​are first averaged to obtain a global first average value. Then, based on the first average value, the audio signal values ​​in the first audio signal sequence are divided. Audio signal values ​​greater than the first average value are averaged to obtain a second average value, and audio signal values ​​less than the first average value are averaged to obtain a third average value. Audio signal values ​​greater than the second average value are assigned to the second average value, and audio signal values ​​less than the third average value are assigned to the third average value. Based on this, the upper limit of the obtained second audio signal sequence is the second average value of the first audio signal sequence, and the lower limit is the third average value. The audio signal values ​​fluctuate between the second and third average values, improving the overall concentration of the sequence.

[0065] Figure 5 This is a schematic diagram of the second audio signal sequence provided in an embodiment of the present disclosure, as shown below. Figure 5 As shown, the horizontal axis represents the number of sampling points, and the vertical axis represents the normalization result. Figure 4 The first audio signal sequence shown is [x0, x1, ..., x...]. n Based on this, we first perform overall mean averaging to obtain the first average value. all and for [x0,x1,…,x n The middle value is greater than the mean. all The audio signal value determines the second average mean. up and less than mean all The audio signal value determines the third average mean. down And let [x0,x1,…,x] n [The value is greater than the mean] up The audio signal value is equal to mean up Let [x0,x1,…,x] n [The middle is less than the mean] down The audio signal value is equal to mean down To obtain the second audio signal sequence as follows Figure 5 As shown, by employing symmetrical mean filtering, the largest and smallest peaks in the audio signal sequence are removed, thereby further concentrating the resulting second audio signal sequence.

[0066] Step 203: Sort the audio signal values ​​in the second audio signal sequence to obtain the third audio signal sequence.

[0067] In this embodiment of the disclosure, the audio signal values ​​of the second audio signal sequence can be sorted. The sorting can be based on the descending order of the audio signal values, so that the obtained third audio signal sequence is monotonically decreasing; or it can be based on the ascending order of the audio signal values, so that the obtained third audio signal sequence is monotonically increasing. Those skilled in the art can choose the sorting method of the audio signal values ​​in the second audio signal sequence according to the data processing conditions, subsequent task requirements, etc., and this embodiment of the disclosure does not impose specific limitations on this.

[0068] In one optional embodiment of the method disclosed herein, step 203 may specifically include step B as follows.

[0069] Step B: Sort the audio signal values ​​in the second audio signal sequence in descending order of numerical value to obtain the third audio signal sequence.

[0070] In this method, the audio signal values ​​in the second audio signal sequence can be sorted in descending order. By sorting the audio signal values ​​in the second audio signal sequence from largest to smallest, the audio signal values ​​in the obtained third audio signal sequence will show a monotonically decreasing trend. That is, the value of each audio signal value is greater than or equal to the value of the next adjacent audio signal value, thus intuitively representing the change in the energy distribution of the audio signal.

[0071] Figure 6 This is a schematic diagram of the third audio signal sequence provided in an embodiment of the present disclosure, as shown below. Figure 6 As shown, the horizontal axis represents the number of sampling points, and the vertical axis represents the normalization result. Figure 5 Based on the second audio signal sequence shown, sort in descending order to obtain the following: Figure 6 The third audio signal sequence f(x) shown has a sequence length N of 10000, and the audio signal values ​​are distributed in the interval [0.45, 0.39]. The audio signal values ​​decrease monotonically in the direction of increasing sequence length.

[0072] Step 204: Determine the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence.

[0073] In this embodiment, the third audio signal sequence can reflect the overall distribution and trend of audio signal intensity, and the rate of change between adjacent audio signal values ​​can reflect the changes in signal intensity. Based on this, according to the extraction requirements of the video segment, a segmentation threshold can be determined based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence. This allows the segmentation threshold to better distinguish the amplitude fluctuations from the foreground to the background and from the background to the foreground in the target video, and to determine the segmentation threshold based on the distribution characteristics of the audio signal values.

[0074] Therefore, based on step B, step 204 may include the following steps C1 to C2.

[0075] Step C1: Determine the difference corresponding to each audio signal value in the third audio signal sequence. The difference is obtained by subtracting each audio signal value from its adjacent audio signal values ​​along the sorting direction.

[0076] Step C2: When the difference meets the rate of change of the second average value relative to the third average value, the audio signal value corresponding to the difference is used as the segmentation threshold.

[0077] In this embodiment of the disclosure, the difference corresponding to each audio signal value in the third audio signal sequence can be determined. This difference is obtained by subtracting the corresponding audio signal value from the adjacent audio signal values ​​in the sorting direction. For example, if the audio signal value of the first position in the third audio signal sequence is x1, the audio signal value of the second position is x2, and the audio signal value of the third position is x3, then the difference corresponding to x1 is obtained by subtracting x1 from x2, the difference corresponding to x2 is obtained by subtracting x2 from x3, and so on. This difference can be used as the derivative corresponding to the audio signal value, representing the rate of change of the audio signal value.

[0078] Furthermore, since the second average value can characterize the overall properties of audio signal values ​​in the first audio signal sequence that are greater than the first average value, and the third average value can characterize the overall properties of audio signal values ​​in the first audio signal sequence that are less than the first average value, the rate of change of the second average value relative to the third average value can characterize the overall trend of change of audio signal values ​​in the first audio signal sequence. The overall audio signal values ​​in the first audio signal sequence change significantly from the foreground to the background and from the background to the foreground. Based on this, when the difference between the audio signal values ​​matches the rate of change of the second average value relative to the third average value, it can be determined that the audio signal value can distinguish the trend of change of the first audio signal sequence from the background to the foreground and from the foreground to the background, thus allowing the audio signal value to be used as a segmentation threshold. In this scheme, by analyzing the distribution characteristics of the audio signal itself, a segmentation threshold that can better distinguish between the foreground and background is obtained. Taking reading video in liveness detection as an example, without introducing hyperparameters, a segmentation threshold that distinguishes reading segments from background segments can be obtained based on the analysis of the audio signal distribution characteristics.

[0079] For example, along the sorting direction of f(x), when the difference between each audio signal value f(i) and the next adjacent audio signal value conforms to the following formula (1), the audio signal value is used as the segmentation threshold:

[0080]

[0081] Where N is 10000, when the audio signal value f(i) corresponding to the i-th sampling point conforms to the above formula (1) with the adjacent audio signal value f(i+1) of the (i+1)-th position, the output f(i) is used as the segmentation threshold t.

[0082] In one optional embodiment of the method disclosed herein, the location information includes a start position and an end position.

[0083] In this embodiment of the disclosure, the position information may be the position in the second audio signal sequence that divides the foreground part and the background part, and may include the start position and the end position of the foreground part in the first audio signal sequence, with the foreground part between the start position and the end position being the foreground part, and the background part between the start position and the end position being the background part.

[0084] Step 205: Along the temporal direction of the first audio signal sequence, determine the position corresponding to the first sampling point as the starting position, where the first sampling point is the first audio signal value greater than or equal to the segmentation threshold; and determine the position corresponding to the second sampling point as the ending position, where the second sampling point is the last audio signal value greater than or equal to the segmentation threshold.

[0085] In this embodiment of the disclosure, the position information can be determined from the signal strength and intensity change trend of the audio signal using the segmentation threshold. For example, the position in the second audio signal sequence where the audio signal value increases from less than the segmentation threshold to greater than the segmentation threshold in the temporal direction can be taken as the starting position. The increase in audio signal strength at the starting position can indicate the start of the user reading the target content. The position where the audio signal value decreases from greater than the segmentation threshold to less than the segmentation threshold can be taken as the ending position. The decrease in audio signal strength at the ending position can indicate the end of the user reading the target content.

[0086] In an optional embodiment of the method disclosed herein, the starting position can be the position corresponding to a first sampling point, which is the first audio signal value greater than or equal to a segmentation threshold. The ending position can be the position corresponding to a second sampling point, which is the last audio signal value greater than or equal to the segmentation threshold. Specifically, audio signal values ​​greater than the segmentation threshold can be assigned a first standard value, and audio signal values ​​less than the segmentation threshold can be assigned a second standard value, wherein the first standard value and the second standard value are different. The first standard value can indicate that the sampling point belongs to the foreground portion, and the second standard value can indicate that the sampling point belongs to the background portion. Thus, in the temporal direction of the second audio signal sequence, the starting position of the foreground portion can be determined based on the position of the first standard value, and the ending position of the foreground portion can be determined based on the position of the last first standard value.

[0087] Figure 7 This is a schematic diagram illustrating the location information determination provided in this embodiment of the disclosure. The horizontal axis represents the number of sampling points. When the first standard value is set to 1 and the second standard value is set to 0, the location information determination is based on the first audio signal sequence [x0, x1, ..., x...]. n In this context, audio signal values ​​greater than the segmentation threshold can be assigned a value of 1, and audio signal values ​​less than the segmentation threshold can be assigned a value of 0. For example... Figure 7As shown. At this point, we can move the position x of the first "1" along the time sequence direction, that is, along the direction in which the sampling points increase from 0 to 10000. p Determine the starting position and set the position of the last "1" to x. q The termination position has been determined.

[0088] Step 206: Deframe the target video to obtain a video frame sequence. The video frame sequence corresponds to the first audio signal sequence in the temporal direction.

[0089] In this embodiment, the target video can be deframed to obtain a video frame sequence. This video frame sequence has temporal information and corresponds to a first audio signal sequence in the temporal direction. Specifically, each sampling point in the first audio signal sequence corresponds to a video frame in the video frame sequence. Deframed video can be implemented using image processing libraries such as OpenCV. Furthermore, deframed video can be performed simultaneously with the extraction of the first audio signal sequence, or it can be performed after determining the location information corresponding to the user reading the target content. Those skilled in the art can choose the method for deframed video according to actual needs, and can also choose the order of steps for obtaining the video frame sequence from the target video based on data processing conditions, detection efficiency requirements, etc. This embodiment does not impose specific limitations in this regard.

[0090] Step 207: Determine the starting frame in the temporal direction of the video frame sequence based on the starting position.

[0091] Step 208: Determine the termination frame in the temporal direction of the video frame sequence based on the termination position.

[0092] In this embodiment of the disclosure, based on the characteristic that the first audio signal sequence and the target implementation correspond in time, the corresponding start frame can be determined in the time direction of the video frame sequence according to the start position in the first audio signal sequence, and the corresponding end frame can be determined in the time direction of the video frame sequence according to the end position in the first audio signal sequence.

[0093] For example, taking a video frame sequence that includes frame F as an example, the starting frame f1 can be determined by the following formula (2):

[0094]

[0095] Furthermore, the termination frame f2 can be determined using the following formula (3):

[0096]

[0097] Step 209: Based on the start frame and the end frame, determine the video segment corresponding to the target content read by the user in the video frame sequence.

[0098] Furthermore, based on the determination of the start frame and the end frame, a video segment between the start frame and the end frame can be determined in the video frame sequence. This video segment is the foreground part of the target content read by the user in the target video, so that the determined video segment corresponds to the audio signal of the foreground part in the first audio signal sequence. Taking the reading video collected by liveness detection as an example, the extracted video segment corresponds to the reading segment of the object to be tested in the audio signal sequence.

[0099] In one optional embodiment of the method disclosed herein, the determined video segments can be used for subsequent detection tasks, such as action detection, face detection, forgery detection, etc. in liveness detection. By accurately and efficiently extracting video segments, the efficiency of the overall detection task and the accuracy of the subsequent output detection results can be improved.

[0100] In the video segment determination method provided in this disclosure, a first audio signal sequence can be extracted from the target video, wherein the target video is the video captured when the user reads the target content, and the first audio signal sequence includes the audio signal values ​​corresponding to each sampling point; further, a segmentation threshold is determined based on the first audio signal sequence, the segmentation threshold being the critical value of the audio signal corresponding to the sampling point belonging to the video segment of the target content being read; then, the video segment corresponding to the target content read by the user is determined based on the segmentation threshold and the first audio signal sequence. In this method, the video segments in the target video are determined based on the analysis of the audio signal, utilizing the temporal correspondence between audio and video in the target video. It does not rely on image processing technology, does not require complex algorithm structures, is simple to implement, has low requirements for video quality, and thus has low requirements for hardware deployment such as recording, shooting, and transmission, and has good versatility; at the same time, the segmentation threshold is determined based on the distribution of audio signal values ​​in the audio sequence, without the need for manual introduction or specification of hyperparameters, and parameter tuning is performed in different scenarios, resulting in unified segmentation rules, low extraction cost, and good robustness.

[0101] Figure 8 A flowchart of the steps of the liveness detection method provided in the embodiments of this disclosure is shown below. Figure 8 As shown, the method may include the following steps 801 to 803.

[0102] Step 801: Obtain the target video of the user to be tested reading the target content.

[0103] In this embodiment of the disclosure, the user to be detected can be a user undergoing liveness detection. The target content to be read can be prompted to the user on-site, and the user's facial movements can be filmed during the reading process. For example, random numerical content or text content can be prompted to the user, and video of the user's facial movements while reading the numerical or text content can be filmed. Filming can be stopped after receiving the user's end command or after the filming time reaches a preset time, and the target video of the user can be obtained. Alternatively, the target video can be obtained from the cloud or from the target video filmed on-site by other terminals. This embodiment of the disclosure does not specifically limit the source of the target video.

[0104] Step 802, using the aforementioned Figures 1 to 7 The method identifies the video segment corresponding to when the user reads the target content from the target video.

[0105] In this embodiment of the disclosure, the aforementioned methods may be used. Figures 1 to 7 The video segment determination method described herein identifies the video segment corresponding to when the user reads the target content from the target video. The specific method steps and flow can be described in the parameter-related description, and will not be repeated here to avoid repetition.

[0106] Step 803: Perform liveness detection on the user to be detected based on the video clip.

[0107] In this embodiment of the disclosure, based on the video segment determined in the target video, liveness detection can be performed on the user to be tested based on the video segment. Specifically, it can include facial forgery detection and lip movement change detection of the user to be tested, so as to determine whether the user to be tested is a live person or whether the user to be tested has correctly read the target content, etc.

[0108] In the liveness detection method provided in this disclosure, a target video of the user reading the target content can be obtained, and then the aforementioned methods can be used. Figures 1 to 7 The video segment determination method identifies video segments within a target video where the user reads target content, and performs liveness detection on the user to be tested based on these video segments. This method utilizes the aforementioned... Figures 1 to 7 The method described above determines video segments by utilizing the temporal correspondence between audio and video in the target video. Based on the analysis of audio signals, it identifies video segments within the target video. It does not rely on image processing technology, does not require complex algorithms, is simple to implement, and has low requirements for video quality. Consequently, it has low requirements for hardware deployment in recording, shooting, and transmission, and is highly versatile. Furthermore, the segmentation threshold is determined based on the distribution of audio signal values ​​in the audio sequence, eliminating the need for manual introduction or specification of hyperparameters. Parameters can be adjusted in different scenarios, resulting in unified segmentation rules, low extraction costs, and good robustness.

[0109] Figure 9 This is a schematic diagram of the structure of the video segment determination device 900 provided in an embodiment of this disclosure. Figure 9 As shown, the device may include: an audio extraction module 901, used to extract a first audio signal sequence from a target video; wherein the target video is a video captured when a user reads target content, and the first audio signal sequence includes audio signal values ​​corresponding to each sampling point; a threshold determination module 902, used to determine a segmentation threshold based on the first audio signal sequence; wherein the segmentation threshold is the audio signal threshold corresponding to a video segment where the sampling point belongs to the video segment where the target content is read; a position determination module 903, used to determine the position information of the sampling point corresponding to the user reading the target content based on the segmentation threshold and the first audio signal sequence; and a segment determination module 904, used to determine the video segment corresponding to the target content read by the user from the target video based on the position information.

[0110] In one optional embodiment of the present disclosure, the threshold determination module 902 is specifically used to filter the first audio signal sequence to obtain a second audio signal sequence; sort the audio signal values ​​in the second audio signal sequence to obtain a third audio signal sequence; and determine the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence.

[0111] In an optional device embodiment of this disclosure, the position information includes a start position and an end position. The position determination module 903 is specifically used to determine the position corresponding to the first sampling point as the start position along the temporal direction of the first audio signal sequence, wherein the first sampling point is the first audio signal value greater than or equal to the segmentation threshold; and to determine the position corresponding to the second sampling point as the end position, wherein the second sampling point is the last audio signal value greater than or equal to the segmentation threshold.

[0112] In an optional device embodiment of this disclosure, the threshold determination module 902 is further configured to perform averaging processing on the audio signal values ​​in the first audio signal sequence to obtain a first average value; perform averaging processing on the audio signal values ​​in the first audio signal sequence that are greater than the first average value to obtain a second average value; and perform averaging processing on the audio signal values ​​in the first audio signal sequence that are less than the first average value to obtain a third average value; assign the audio signal values ​​in the first audio signal sequence that are greater than the second average value to the second average value, and assign the audio signal values ​​that are less than the third average value to the third average value to obtain a second audio signal sequence.

[0113] In an optional device embodiment of this disclosure, the threshold determination module 902 is further configured to sort the audio signal values ​​in the second audio signal sequence in descending order of value to obtain a third audio signal sequence; determine the difference corresponding to each audio signal value in the third audio signal sequence, the difference being obtained by subtracting each audio signal value from its adjacent audio signal value along the sorting direction; and when the difference meets the rate of change of the second average value relative to the third average value, use the audio signal value corresponding to the difference as a segmentation threshold.

[0114] In an optional device embodiment of this disclosure, the segment determination module 904 is further configured to deframe the target video to obtain a video frame sequence, wherein the video frame sequence corresponds to the first audio signal sequence in the temporal direction; determine a start frame in the temporal direction of the video frame sequence according to the start position; determine an end frame in the temporal direction of the video frame sequence according to the end position; and determine the video segment corresponding to the target content read by the user in the video frame sequence according to the start frame and the end frame.

[0115] In the video segment determination device provided in this disclosure, a first audio signal sequence can be extracted from the target video, wherein the target video is the video captured when the user reads the target content, and the first audio signal sequence includes the audio signal values ​​corresponding to each sampling point; further, a segmentation threshold is determined based on the first audio signal sequence, the segmentation threshold being the critical value of the audio signal corresponding to the sampling point belonging to the video segment of the target content being read; then, the video segment corresponding to the target content read by the user is determined based on the segmentation threshold and the first audio signal sequence. In this method, the video segments in the target video are determined based on the analysis of the audio signal, utilizing the temporal correspondence between audio and video in the target video. It does not rely on image processing technology, does not require complex algorithm structures, is simple to implement, has low requirements for video quality, and thus has low requirements for hardware deployment such as recording, shooting, and transmission, and has good versatility; at the same time, the segmentation threshold is determined based on the distribution of audio signal values ​​in the audio sequence, without the need for manual introduction or specification of hyperparameters, and parameter tuning is performed in different scenarios, resulting in unified segmentation rules, low extraction cost, and good robustness.

[0116] Figure 10 A structural block diagram of a liveness detection device 1000 provided in this embodiment of the present disclosure is shown below. Figure 10 As shown, the device may include: a video acquisition module 1001, used to acquire a target video of a user reading target content; and a segment determination module 1002, used to employ methods such as... Figure 9 The device determines the video segment corresponding to when the user to be detected reads the target content from the target video; the liveness detection module 1003 is used to perform liveness detection on the user to be detected based on the video segment.

[0117] The liveness detection device provided in this disclosure can acquire a target video of the user reading the target content, and then the aforementioned methods can be used. Figure 9 The video segment determination device determines a video segment within a target video where the user reads target content, and performs liveness detection on the user to be tested based on this video segment. This method utilizes the aforementioned... Figure 9 The device identifies video segments by utilizing the temporal correspondence between audio and video in the target video. It determines video segments in the target video based on the analysis of audio signals. It does not rely on image processing technology, does not require complex algorithms, is simple to implement, and has low requirements for video quality. Therefore, it has low requirements for hardware deployment such as recording, shooting, and transmission, and has good versatility. At the same time, the segmentation threshold is determined based on the distribution of audio signal values ​​in the audio sequence, without the need for manual introduction or specification of hyperparameters. The parameters can be adjusted in different scenarios, resulting in unified segmentation rules, low extraction costs, and good robustness.

[0118] Figure 11 This is a schematic diagram of the structure of an electronic device 1100 provided in an embodiment of this disclosure, as shown below. Figure 11 As shown, the electronic device 1100 may include a processor 1101, a memory 1102, and a program or instructions stored in the memory 1102 and executable on the processor 1101. When the program or instructions are executed by the processor 1101, they implement the various processes of the above-described video segment determination method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0119] It should be noted that, Figure 11 The illustrated electronic device 1100 is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0120] Figure 12 This is a hardware schematic diagram of an electronic device 1200 provided in an embodiment of the present disclosure, as shown below. Figure 12 As shown, the electronic device 1200 includes a Central Processing Unit (CPU) 1201, which can perform various appropriate actions and processes based on programs stored in ROM (Read Only Memory) 1202 or programs loaded from storage section 1208 into RAM (Random Access Memory) 1203. The RAM 1203 also stores various programs and data required for system operation. The CPU 1201, ROM 1202, and RAM 1203 are interconnected via bus 1204. An I / O (Input / Output) interface 1205 is also connected to bus 1204.

[0121] The following components are connected to I / O interface 1205: an input section 1206 including a keyboard, mouse, etc.; an output section 1207 including a CRT (Cathode Ray Tube), LCD (Liquid Crystal Display), etc., and speakers, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 1209 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to I / O interface 1205 as needed. Removable media 1211, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 1210 as needed so that computer programs read from them can be installed into storage section 1208 as needed.

[0122] In particular, according to embodiments of this disclosure, the processes described below with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this 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 such embodiments, the computer program can be downloaded and installed from a network via communication section 1209, and / or installed from removable medium 1211. When the computer program is executed by the central processing unit (CPU 1201), it performs various functions defined in the system of this application.

[0123] This disclosure also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described video segment determination method or liveness detection method embodiment, and achieve the same technical effect. To avoid repetition, further details are omitted here.

[0124] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media such as ROM, RAM, magnetic disk, or optical disk.

[0125] This disclosure also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described video segment determination method or liveness detection method embodiments, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0126] It should be understood that the chip mentioned in the embodiments of this disclosure may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0127] This disclosure provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the steps of the video segment determination method or the liveness detection method embodiment described above, and achieve the same technical effect. To avoid repetition, further details are omitted here.

[0128] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this disclosure is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0129] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal (which may be a mobile phone, computer, electronic device, air conditioner, or network device, etc.) to execute the methods of the various embodiments of this disclosure.

[0130] The embodiments of this disclosure have been described above with reference to the accompanying drawings. However, this disclosure is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this disclosure without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this disclosure.

Claims

1. A method for determining video segments, characterized in that, The method includes: Extract a first audio signal sequence from the target video; wherein the target video is a video captured when the user reads the target content, and the first audio signal sequence includes the audio signal values ​​corresponding to each sampling point; A segmentation threshold is determined based on the first audio signal sequence; wherein, the segmentation threshold is the critical value of the audio signal corresponding to the case where the sampling point belongs to a video segment of the target content; Based on the segmentation threshold and the first audio signal sequence, determine the location information of the sampling point corresponding to when the user reads the target content; Based on the location information, determine the video segment corresponding to the user's target content from the target video; The step of determining the segmentation threshold based on the first audio signal sequence includes: filtering the first audio signal sequence to obtain a second audio signal sequence; sorting the audio signal values ​​in the second audio signal sequence to obtain a third audio signal sequence; and determining the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence. The step of determining the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence includes: determining the difference corresponding to each audio signal value in the third audio signal sequence, wherein the difference is obtained by subtracting each audio signal value from its adjacent audio signal values ​​along the sorting direction; when the difference meets the rate of change of the second average value relative to the third average value, the audio signal value corresponding to the difference is used as the segmentation threshold; wherein the second average value is obtained by averaging the audio signal values ​​in the first audio signal sequence that are greater than the first average value; the third average value is obtained by averaging the audio signal values ​​in the first audio signal sequence that are less than the first average value; and the first average value is obtained by averaging the audio signal values ​​in the first audio signal sequence.

2. The method according to claim 1, characterized in that, The location information includes a start position and an end position. Determining the location information of the sampling point corresponding to when the user reads the target content, based on the segmentation threshold and the first audio signal sequence, includes: Along the temporal direction of the first audio signal sequence, the position corresponding to the first sampling point is determined as the starting position, where the first sampling point is the first audio signal value that is greater than or equal to the segmentation threshold. Furthermore, the position corresponding to the second sampling point is determined as the termination position, where the second sampling point is the sampling point where the last audio signal value is greater than or equal to the segmentation threshold.

3. The method according to claim 1, characterized in that, The step of filtering the first audio signal sequence to obtain the second audio signal sequence includes: The audio signal values ​​in the first audio signal sequence are averaged to obtain a first average value. The audio signal values ​​in the first audio signal sequence that are greater than the first average value are averaged to obtain a second average value, and the audio signal values ​​in the first audio signal sequence that are less than the first average value are averaged to obtain a third average value. The audio signal values ​​in the first audio signal sequence that are greater than the second average value are assigned the second average value, and the audio signal values ​​that are less than the third average value are assigned the third average value, thereby obtaining the second audio signal sequence.

4. The method according to claim 1, characterized in that, The step of sorting the audio signal values ​​in the second audio signal sequence to obtain the third audio signal sequence includes: The audio signal values ​​in the second audio signal sequence are sorted in descending order to obtain the third audio signal sequence; Accordingly, determining the segmentation threshold based on the rate of change between adjacent audio signal values ​​in the third audio signal sequence includes: Determine the difference corresponding to each audio signal value in the third audio signal sequence, wherein the difference is obtained by subtracting each audio signal value from its adjacent audio signal values ​​along the sorting direction; When the difference meets the rate of change of the second average value relative to the third average value, the audio signal value corresponding to the difference is used as the segmentation threshold.

5. The method according to claim 2, characterized in that, The step of determining the video segment corresponding to the user's target content from the target video based on the location information includes: The target video is deframed to obtain a video frame sequence, which corresponds to the first audio signal sequence in the temporal direction. Based on the starting position, the starting frame is determined in the temporal direction of the video frame sequence; Based on the termination position, a termination frame is determined in the temporal direction of the video frame sequence; Based on the start frame and the end frame, the video segment corresponding to the user's target content is determined in the video frame sequence.

6. A method for detecting liveness, characterized in that, The method includes: Obtain the target video of the user being tested reading the target content; Using the method described in any one of claims 1-5, determine from the target video the video segment corresponding to when the user to be detected reads the target content; Liveness detection is performed on the user to be detected based on the video clip.

7. An electronic device, characterized in that, The electronic device includes: Processor; and Memory for storing the executable instructions of the processor; The processor is configured to execute the video segment determination method of any one of claims 1-5 or the liveness detection method of claim 6 by executing the executable instructions.

8. A computer-readable storage medium storing a computer program thereon, characterized in that, When the computer program is executed by the processor, it implements the video segment determination method according to any one of claims 1-5, or the liveness detection method according to claim 6.

9. A computer program product, characterized in that, When the computer program product is run on an electronic device, it causes the electronic device to perform the video segment determination method as described in any one of claims 1-5, or the liveness detection method as described in claim 6.

Images