Video quality assessment method and apparatus, device, medium, and product

Abstract

The present application relates to a video quality assessment method and apparatus, a device, a medium, and a product. The method comprises: acquiring a target video, a preset video quality objective assessment model, and preset subjective assessment level categories; on the basis of the subjective assessment level categories, determining a first value range for the optimization of a first model parameter of the video quality objective assessment model and a second value range for the optimization of a second model parameter of the video quality objective assessment model; and acquiring subjective assessment level results of multiple users for the target video, and on the basis of the first value range, the second value range, and the subjective assessment level results, continuously and iteratively correcting the first mode parameter and the second model parameter until the output accuracy of the corrected video quality objective assessment model meets a preset accuracy threshold, so as to obtain a target video quality assessment model, thereby determining a quality score of the target video. The method can improve the accuracy and reliability of objectively assessing the quality of a target video.

Description

Video quality assessment methods, devices, equipment, media and products Technical Field

[0001] This application relates to the technical field of video processing, and in particular to a video quality evaluation method, apparatus, device, medium, and product. Background Technology

[0002] Quality of Experience (QoE) is a service evaluation method based on user satisfaction. It integrates factors from the service, user, and environmental levels, and directly reflects the degree of user satisfaction with the service.

[0003] In related technologies, video quality assessment methods are mainly divided into two types: subjective quality assessment and objective quality assessment. Subjective quality assessment involves users directly rating the video quality. This method is the most accurate, but it is labor-intensive and time-consuming, making it unsuitable for large-scale applications. Objective quality assessment methods use algorithms or models to quantify video quality, primarily considering network performance such as throughput, latency, packet loss rate, jitter, and bit error rate. Although automation is possible, its accuracy is generally lower than subjective evaluation. In other words, the reliability of existing objective video quality assessment methods is not high. Summary of the Invention

[0004] Therefore, it is necessary to address the technical problem of low reliability of existing objective video quality evaluation methods by considering a combination of subjective and objective approaches, and to provide a video quality evaluation method, apparatus, computer equipment, computer-readable storage medium, and computer program product.

[0005] Firstly, this application provides a video quality evaluation method, which includes:

[0006] The system acquires the target video, a preset objective video quality evaluation model, and a preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective video quality evaluation model is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0007] Based on the subjective evaluation level classification, determine the first value range for optimizing the first model parameters, and based on the subjective evaluation level classification, determine the second value range for optimizing the second model parameters.

[0008] Obtain the subjective evaluation results of multiple users for the target video. Based on the first value range, the second value range, and the subjective evaluation results, continuously iterate and correct the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0009] Based on the target video quality evaluation model, determine the quality score of the target video.

[0010] In one embodiment, determining a first value range for optimizing the first model parameters based on the subjective evaluation level classification includes:

[0011] Obtain the preset mapping table between subjective evaluation level categories and evaluation scores, and determine the highest evaluation score corresponding to the highest evaluation level and the lowest evaluation score corresponding to the lowest evaluation level.

[0012] The upper limit of the first model parameter is determined based on the highest evaluation score, and the lower limit of the first model parameter is determined based on the lowest evaluation score.

[0013] The first value range is determined based on the lower limit and the upper limit of the first model parameter.

[0014] In one embodiment, the second value range for optimizing the second model parameters is determined based on the subjective evaluation level classification, including:

[0015] Obtain the maximum packet loss rate of the wide area network where the target video is located, and determine the initial value range of the second model parameters based on the maximum packet loss rate;

[0016] The transmission packet loss rate is determined as the independent variable parameter of the objective video quality evaluation model. Sampling is carried out within the initial value range to obtain sampled data. The sampled data includes the network packet loss rate value and the evaluation value obtained after the network packet loss rate value is input into the objective video quality evaluation model.

[0017] Based on each evaluation value, the upper limit of the second model parameter is determined, and the initial value range is updated based on the upper limit of the second model parameter to obtain the second value range.

[0018] In one embodiment, the first model parameters and the second model parameters are iteratively corrected based on a first value range and a second value range until the output accuracy of the corrected video quality objective evaluation model meets a preset accuracy threshold, including:

[0019] Based on the first value range and the second value range, the first initial value of the first model parameter and the second initial value of the second model are determined respectively;

[0020] The first and second initial values ​​are iterated a preset number of times with equal step size to obtain multiple first quality score output values. Based on the first quality score output values, the updated first and second initial values ​​are determined.

[0021] The updated first and second initial values ​​are iterated for a preset number of steps to obtain multiple second quality score output values. Based on the second quality score output values, the updated first and second initial values ​​are further optimized until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold.

[0022] In one embodiment, determining the updated first initial value and the updated second initial value based on the first quality score output value includes:

[0023] Based on the output values ​​of each first quality score, determine the corresponding network impairment value, obtain the subjective evaluation value of at least one user based on the network impairment value, and determine the average value of the subjective evaluation value.

[0024] Obtain a preset deviation optimization function, input the average value of the corresponding network damage value and the subjective evaluation value into the deviation optimization function to obtain the deviation value, determine the first initial value corresponding to the minimum deviation value as the updated first initial value, and determine the second initial value corresponding to the minimum variation value as the updated second initial value.

[0025] In one embodiment, based on a first value range, a second value range, and the subjective evaluation level result, the first model parameters and the second model parameters are iteratively corrected until the output accuracy of the corrected video quality objective evaluation model meets a preset accuracy threshold, including:

[0026] Based on the subjective evaluation results, key points and their corresponding actual quality scores are determined. Key points are used to characterize the abrupt changes in video quality evaluation levels.

[0027] Based on the first and second value ranges, the first and second model parameters are iteratively corrected until the deviation between the output value of the corrected video quality objective evaluation model and the actual quality score value of the corresponding key point is not greater than the preset deviation.

[0028] Secondly, this application also provides a video quality evaluation device, which includes:

[0029] The evaluation data acquisition module is used to acquire the target video, the preset objective evaluation model for video quality, and the preset subjective evaluation level classification. The first model parameter in the objective evaluation model for video quality is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective evaluation model for video quality is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0030] The model parameter optimization module is used to determine the first value range for optimizing the first model parameter based on the subjective evaluation level classification, and to determine the second value range for optimizing the second model parameter based on the subjective evaluation level classification.

[0031] The evaluation model generation module is used to obtain the subjective evaluation level results of multiple users for the target video. Based on the first value range, the second value range and the subjective evaluation level results, it continuously iterates and corrects the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0032] The video quality assessment module is used to determine the quality score of the target video based on the target video quality assessment model.

[0033] Thirdly, this application also provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0034] The system acquires the target video, a preset objective video quality evaluation model, and a preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective video quality evaluation model is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0035] Based on the subjective evaluation level classification, determine the first value range for optimizing the first model parameters, and based on the subjective evaluation level classification, determine the second value range for optimizing the second model parameters.

[0036] Obtain the subjective evaluation results of multiple users for the target video. Based on the first value range, the second value range, and the subjective evaluation results, continuously iterate and correct the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0037] Based on the target video quality evaluation model, determine the quality score of the target video.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0039] The system acquires the target video, a preset objective video quality evaluation model, and a preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective video quality evaluation model is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0040] Based on the subjective evaluation level classification, determine the first value range for optimizing the first model parameters, and based on the subjective evaluation level classification, determine the second value range for optimizing the second model parameters.

[0041] Obtain the subjective evaluation results of multiple users for the target video. Based on the first value range, the second value range, and the subjective evaluation results, continuously iterate and correct the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0042] Based on the target video quality evaluation model, determine the quality score of the target video.

[0043] Fifthly, this application also provides a computer program product comprising a computer program that, when executed by a processor, performs the following steps:

[0044] The system acquires the target video, a preset objective video quality evaluation model, and a preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective video quality evaluation model is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0045] Based on the subjective evaluation level classification, determine the first value range for optimizing the first model parameters, and based on the subjective evaluation level classification, determine the second value range for optimizing the second model parameters.

[0046] Obtain the subjective evaluation results of multiple users for the target video. Based on the first value range, the second value range, and the subjective evaluation results, continuously iterate and correct the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0047] Based on the target video quality evaluation model, determine the quality score of the target video.

[0048] The aforementioned video quality evaluation method, apparatus, computer equipment, storage medium, and computer program product, when evaluating the quality of a target video, first acquire the target video, a preset objective video quality evaluation model, and a preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model characterizes the impact of network packet loss rate on video quality, and the second model parameter characterizes the degree of correction for the impact of network packet loss rate on video quality. Then, based on the subjective evaluation level classification, a first value range for optimizing the first model parameter is determined, and a second value range for optimizing the second model parameter is determined based on the subjective evaluation level classification. Next, multiple users' subjective evaluation level results for the target video are acquired. Based on the first value range, the second value range, and the subjective evaluation level results, the first and second model parameters are iteratively corrected until the output accuracy of the corrected objective video quality evaluation model meets a preset accuracy threshold, thus obtaining the target video quality evaluation model. Finally, based on the target video quality evaluation model, the quality score of the target video is determined.

[0049] The video quality evaluation method of this application iteratively corrects the first and second model parameters in the preset objective video quality evaluation model based on the user's subjective evaluation level of the target video. This makes the output value of the objective video quality evaluation model continuously approach the user's subjective evaluation level. The corrected objective video quality evaluation model, which finally meets the preset accuracy threshold, is then determined as the target video quality evaluation model, thereby helping to improve the accuracy and reliability of objectively evaluating the quality of the target video. Attached Figure Description

[0050] Figure 1 shows the application environment of a video quality evaluation method in one embodiment;

[0051] Figure 2 is a flowchart of a video quality evaluation method in one embodiment;

[0052] Figure 3 is a flowchart of determining the first value range in one embodiment;

[0053] Figure 4 is a flowchart of determining the second value range in one embodiment;

[0054] Figure 5 is a flowchart of iterating through the first and second initial values ​​in one embodiment;

[0055] Figure 6 is a flowchart of determining the updated second initial value in one embodiment;

[0056] Figure 7 is a flowchart of a modified objective evaluation model for video quality in one embodiment;

[0057] Figure 8 is a schematic diagram of a video quality system in one embodiment;

[0058] Figure 9 is a structural block diagram of a video quality evaluation device in one embodiment;

[0059] Figure 10 is an internal structure diagram of a computer device in one embodiment;

[0060] Figure 11 is an internal structure diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] Quality of Experience (QoE) is a service evaluation method based on user satisfaction. It integrates factors from the service, user, and environmental levels, directly reflecting users' level of satisfaction with the service. Regarding the quantification of QoE, the most widely used method is the "mean opinion score (MOS)," which divides the subjective perception of QoE into five levels.

[0063] In related technologies, video quality assessment methods are mainly divided into two types: subjective quality assessment and objective quality assessment. Subjective quality assessment involves users directly rating the video quality. This method is the most accurate, but it is labor-intensive and time-consuming, making it unsuitable for large-scale applications. Objective quality assessment methods use algorithms or models to quantify video quality, primarily considering network performance such as throughput, latency, packet loss rate, jitter, and bit error rate. Although automation is possible, its accuracy is generally lower than subjective evaluation. In other words, the reliability of existing objective video quality assessment methods is not high.

[0064] To address the technical problem of low reliability in existing objective video quality evaluation methods, this application proposes a method combining subjective and objective video quality evaluation. This method aims to provide a practical and reliable specific model result while considering network packet loss rate, thereby enabling better evaluation of network video quality. Specifically, the video quality evaluation method provided in this application embodiment can be applied to the application environment shown in Figure 1. The terminal 102 communicates with the server 104 via a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104 or placed on the cloud or other network servers. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers.

[0065] In one embodiment, as shown in Figure 2, taking the application of this method to the terminal in Figure 1 as an example, it can be understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0066] Step 202: Obtain the target video, the preset objective video quality evaluation model, and the preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective video quality evaluation model is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0067] Generally, video quality is mainly reflected in three aspects: clarity, smoothness, and latency. Video clarity mainly reflects the video resolution during playback; video smoothness mainly reflects the stuttering during playback; and video latency mainly reflects the transitions between live streams and the start of video-on-demand playback. The user's subjective perception is highly sensitive to clarity and smoothness, and these are major factors influencing subjective ratings. Furthermore, data packet loss occurs during network transmission, often impacting both clarity and smoothness. Extensive experiments have shown that video playback quality exhibits a non-linear relationship with network packet loss, and the quality declines rapidly. Additionally, the QoE (Quality of Experience) rating for video quality assessment is typically on a 5-point scale (0-5).

[0068] The target video is used to characterize the video to be evaluated for quality. The objective video quality evaluation model can be preset based on the type of the target video and the evaluation requirements. The preset subjective evaluation level classification is typically five categories: Excellent, Good, Average, Poor, and Very Poor, with corresponding score values ​​of 5, 4, 3, 2, and 1. The first and second model parameters are both undetermined coefficients of the preset objective video quality evaluation model. Network packet loss rate refers to the proportion of data packets lost during network transmission out of the total number of data packets sent. It is usually expressed as a percentage (%) or per mille (‰). Network packet loss rate directly affects the quality of video transmission; a high packet loss rate can lead to video stuttering, blurring, and latency, thus reducing the user experience.

[0069] For example, with a preset subjective evaluation level classification of five levels: very good, good, average, poor, and very poor, the preset objective evaluation model for video quality uses the following objective evaluation model calculation formula as an example: y(x)=5 / [m (n*x) ];

[0070] In the formula, y(x) represents the model output value; m represents the first model parameter; and n represents the second model parameter. In practical applications, the impact of packet loss rate on video quality is usually not a simple linear relationship. Instead, as the packet loss rate increases, the rate of video quality degradation may vary. The first model parameter m, by adjusting the nonlinear relationship in the model, can more accurately reflect this impact. Within a certain range of packet loss rates, the impact of packet loss rate on video quality may exhibit a specific trend. The second model parameter n is used to correct this trend, ensuring that the model provides accurate assessments across different packet loss rate ranges. The combined effect of the first and second model parameters improves the accuracy and reliability of the final fitting result.

[0071] Step 204: Based on the subjective evaluation level classification, determine the first value range for optimizing the first model parameters, and based on the subjective evaluation level classification, determine the second value range for optimizing the second model parameters.

[0072] To reflect users' subjective experience and improve the model's adaptability, accuracy, robustness, and reliability, the first value range for optimizing the first model parameters and the second value range for optimizing the second model parameters are determined based on the subjective evaluation level classification.

[0073] Specifically, as can be seen from the above, since the preset subjective evaluation level classification in this embodiment corresponds to the score values ​​of 5, 4, 3, 2 and 1 respectively, in order to ensure that the output value of the finally optimized model remains between [1,5], it is necessary to determine the value range of the first model parameter and the value range of the second model parameter.

[0074] Step 206: Obtain subjective evaluation results of multiple users for the target video. Based on the first value range, the second value range, and the subjective evaluation results, continuously iterate and correct the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0075] To accurately integrate user subjective experience and objective video evaluation, and ultimately obtain a highly reliable video quality evaluation model, the parameters of the first and second models need to be iteratively corrected. In each iteration, the model parameters (first and second model parameters) are adjusted, and the error between the model output and the subjective evaluation result is calculated. The accuracy of the model is evaluated by calculating this error. When the accuracy of the model output meets a preset accuracy threshold, the iterative optimization process stops, and the final target video quality evaluation model is obtained.

[0076] Step 208: Determine the quality score of the target video based on the target video quality evaluation model.

[0077] After obtaining the target video quality evaluation model, the quality score of the target video can be determined based on the actual packet loss rate, and the reliability of this quality score is relatively high.

[0078] In the aforementioned video quality evaluation method, the first and second model parameters in the preset objective video quality evaluation model are iteratively corrected based on the user's subjective evaluation level of the target video. This makes the output value of the objective video quality evaluation model continuously approach the user's subjective evaluation level. The corrected objective video quality evaluation model, whose final output accuracy meets the preset accuracy threshold, is then determined as the target video quality evaluation model. This helps to improve the accuracy and reliability of objectively evaluating the quality of the target video.

[0079] In one embodiment, as shown in Figure 3, the first value range for optimizing the first model parameters is determined based on the subjective evaluation level classification, including:

[0080] Step 302: Obtain the preset mapping relationship table between subjective evaluation level classification and evaluation score, and determine the highest evaluation score corresponding to the highest evaluation level and the lowest evaluation score corresponding to the lowest evaluation level.

[0081] The preset mapping table between subjective evaluation level categories and evaluation scores is used to represent the evaluation scores corresponding to specific subjective evaluation level categories. As can be seen from the above, in this embodiment, the score values ​​(evaluation scores) corresponding to the five subjective evaluation level categories of very good, good, average, poor and very poor are 5, 4, 3, 2 and 1, respectively. That is to say, the highest evaluation score is 5 and the lowest evaluation score is 1.

[0082] Specifically, after determining the highest and lowest evaluation scores, the optimization boundaries of the first and second model parameters in the iterative optimization process are determined.

[0083] Step 304: Determine the upper limit of the first model parameter based on the highest evaluation score, and determine the lower limit of the first model parameter based on the lowest evaluation score.

[0084] Specifically, to ensure that the output value of the final target video quality assessment model is no greater than 5, the value of the first model parameter m must be greater than 1. Therefore, the lower limit of the first model parameter is 1. However, the value of the first model parameter m cannot be too large either, otherwise the output value of the corresponding target video quality assessment model will be too small. Experiments have shown that when the maximum value of the first model parameter m is 10, the output value of the target video quality assessment model will fluctuate between 1 and 5, that is, the upper limit of the first model parameter is 10.

[0085] Step 306: Determine the first value range based on the lower limit and upper limit of the first model parameter.

[0086] For example, according to step 304, the first value range corresponding to the first model parameter is (1, 10).

[0087] In this embodiment, the range of values ​​for the first model parameter is determined based on the user's subjective evaluation of the target video. This clarifies the user's subjective feelings under different video qualities. By associating these subjective feelings with the model parameter, the user's actual experience can be captured more accurately.

[0088] In one embodiment, as shown in Figure 4, the second value range for optimizing the second model parameters is determined based on the subjective evaluation level classification, including:

[0089] Step 402: Obtain the maximum packet loss rate of the wide area network where the target video is located, and determine the initial value range of the second model parameters based on the maximum packet loss rate.

[0090] Specifically, the maximum packet loss rate reflects the worst-case transmission conditions of video in a real-world network. By obtaining this data, it can be ensured that the model parameters can cover the most extreme network conditions during optimization, thereby improving the robustness and adaptability of the model.

[0091] For example, assuming the target video is transmitted in a wide area network (WAN), and measurements show that the maximum packet loss rate of the LAN where the target video resides is 10%, then the initial value of the second model parameter can be determined to be no more than 20. This ensures that the optimization process of the model parameters will not exceed the limitations of the actual network conditions. When the packet loss rate is 10%, the preset model calculation minimum value is y(0.1) = 5 /

[10] .(20*0.1) ], which is 0.05.

[0092] Step 404: The transmission packet loss rate is determined as the independent variable parameter of the objective video quality evaluation model. Sampling is performed within the initial value range to obtain sampled data. The sampled data includes the network packet loss rate value and the evaluation value obtained after inputting the network packet loss rate value into the objective video quality evaluation model.

[0093] Specifically, by sampling within an initial value range, a series of network packet loss rates and their corresponding video quality evaluation values ​​can be obtained. This sampled data provides practical training data for optimizing model parameters, helping the model to be finely adjusted under different network conditions.

[0094] For example, taking the example in step 402, the second model parameter is set within the initial value range [0,20]. Assuming the values ​​are 0%, 2%, 4%, 6%, 8%, and 10%, the corresponding evaluation values ​​are 5.0, 4.8, 4.5, 4.2, 3.8, and 3.5.

[0095] Step 406: Determine the upper limit of the second model parameters based on each evaluation value, and update the initial value range based on the upper limit of the second model parameters to obtain the second value range.

[0096] Specifically, by analyzing the evaluation values ​​in the sampled data, the upper limit of the second model parameters is determined to ensure that the model parameters do not exceed the effective range during the optimization process, avoiding over-optimization or under-optimization. Furthermore, by updating the initial value range, the final value range of the second model parameters is obtained, providing clear targets and constraints for the final optimization of the model.

[0097] For example, taking a maximum packet loss rate of 2% in the wide area network where the target video is located, and setting the initial value of the first model parameter m to 10 and the initial value of the second model parameter to 20 as an example, when the input value of the objective evaluation model for video quality is 0.02, the score calculated by the objective evaluation model for video quality is y(x) = 5 /

[10] (20*x) =1.990535853. Experiments show that the network packet loss rate is relatively reasonable with the model output score. Therefore, there is no need to make the value of n larger. Thus, the upper limit of the second model parameter n can be determined to be 20. That is to say, the second value range of the second model parameter is (0,20).

[0098] In this embodiment, by obtaining the maximum packet loss rate of the wide area network where the target video is located, and determining the initial value range of the second model parameters based on this data, it can be ensured that the model parameters can cover the most extreme network conditions during the optimization process, thereby improving the practical application value, effectiveness, and robustness of the model.

[0099] In one embodiment, as shown in Figure 5, the first model parameters and the second model parameters are iteratively corrected based on the first value range and the second value range until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, including:

[0100] Step 502: Based on the first value range and the second value range, determine the first initial value of the first model parameter and the second initial value of the second model, respectively.

[0101] Specifically, by setting a clear initial value range, the initial values ​​of the model parameters are ensured to be within a reasonable range, avoiding instability in the optimization process caused by initial values ​​that are too large or too small.

[0102] For example, taking the first model parameter m obtained in the above steps as having a value range of (1, 10] and the second model parameter n as having a value range of (0, 20], the initial value of m is 10 and the initial value of n is 20.

[0103] Step 504: Iterate the first and second initial values ​​for a preset number of times with equal step size to obtain multiple first quality score output values. Based on the first quality score output values, determine the updated first initial value and the updated second initial value.

[0104] Specifically, through iterative steps with equal step size, the parameter space is fully explored, and multiple quality score output values ​​are obtained, providing a data foundation for subsequent optimization. In other words, multiple iterations can help discover local optima for the model parameters, providing a reference for subsequent optimization. By analyzing the first quality score output value, the optimization direction is clarified, and the first initial values ​​are updated. Similarly, by analyzing the second quality score output value, the optimization direction is clarified, allowing the model parameters to be adjusted towards a more optimal direction. The step size and the preset number of iterations can be preset based on the output accuracy of the final target video quality evaluation model; in this embodiment, the preset number of iterations can be 100.

[0105] For example, the step size of m is 1, and the step size of n is 2. All permutations are taken, and the optimization is performed approximately one hundred times. The optimization objective is then calculated as MIN T = MIN∑[y(X...]. i )-Y i ] 2 We obtain the first round of optimization points m1 (i.e., the updated first initial value) and n1 (i.e., the updated second initial value), where X i For network packet loss rate; y(X) i ) is the model for input X i The predicted score value, i.e., the first quality score output value; Y i The user's network packet loss rate is X. iThe actual subjective rating value under the given conditions. When the optimization objective is achieved, that is, when the sum of squared differences between the model's predicted rating value and the user's actual subjective rating value is minimized, the updated first initial value m1 and the updated second initial value n1 are obtained.

[0106] Step 506: Iterate the updated first initial value and the updated second initial value for a preset number of iterations with equal step size to obtain multiple second quality score output values. Based on the second quality score output values, continue to optimize the updated first initial value and the updated second initial value until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold.

[0107] Specifically, to further optimize and obtain the final first and second model parameters, based on the updated first initial value (m1) and updated second initial value (n1) obtained in the first round of optimization, it is necessary to continue iterating the updated first initial value and updated second initial value a preset number of times to obtain multiple second quality score output values. Then, based on the second quality score output values, the updated first initial value and updated second initial value are further optimized, that is, according to the optimization objective MIN T=MIN∑[y(X) in step 504 above. i )-Y i ] 2 Continue optimizing until the optimization target reaches its minimum value, that is, the output accuracy of the video quality objective evaluation model of the correction week meets the preset accuracy threshold. At this point, the first initial value obtained by the final update can be determined as the final value of the first model parameter, and the second initial value obtained by the final update is the final value of the second model parameter.

[0108] For example, based on m1 and n1, the step size is reduced by a factor of ten, and in the second round of optimization, the step size of m is 0.1 and the step size of n is 0.2. This is repeated for approximately one hundred optimization calculations, with the goal of minimizing the sum of squared differences. Through a hierarchical approximation optimization algorithm, the y(x) model result is corrected, yielding the final first and second model parameters, thus correcting the objective video quality evaluation model. The corrected model can then be practically applied to video evaluation scenarios. The set-top box probe obtains network packet loss values, which, after being substituted into the model calculation, can provide a quality evaluation value for the video playback.

[0109] In this embodiment, through continuous iterative optimization, the model parameters can be gradually optimized, resulting in a gradual improvement in the model's output accuracy until a preset accuracy threshold is reached. This helps improve the overall performance and reliability of the model, ensuring that the model can provide accurate evaluation results under various network conditions.

[0110] In one embodiment, as shown in Figure 6, determining the updated first initial value and the updated second initial value based on the first quality score output value includes:

[0111] Step 602: Determine the corresponding network impairment value based on each first quality score output value, obtain the subjective evaluation value of at least one user based on the network impairment value, and determine the average value of the subjective evaluation value.

[0112] The first quality score output is the model's predicted score, and the network impairment score is the network packet loss rate corresponding to the model's predicted score. The user's subjective evaluation value is the subjective evaluation value corresponding to the network impairment value.

[0113] Step 604: Obtain the preset deviation optimization function. Input the average value of the one-to-one network damage value and the subjective evaluation value into the deviation optimization function to obtain the deviation value. Determine the first initial value corresponding to the minimum deviation value as the updated first initial value, and determine the second initial value corresponding to the minimum variation value as the updated second initial value.

[0114] Specifically, taking the aforementioned optimization objective as an example, the preset deviation optimization function in this embodiment is the optimization objective: MIN T=MIN∑[y(X i )-Y i ] 2 When the deviation optimization function reaches its minimum value, the first initial value corresponding to the minimum deviation value is determined as the updated first initial value, and the second initial value corresponding to the minimum variation value is determined as the updated second initial value.

[0115] In this embodiment, the updated first initial value and the updated second initial value are determined based on the average of the preset deviation optimization function, network impairment value and subjective evaluation value, which helps to improve the accuracy of updating the first initial value and the second initial value.

[0116] In one embodiment, as shown in Figure 7, the first model parameters and the second model parameters are iteratively corrected based on the first value range, the second value range, and the subjective evaluation level results, until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, including:

[0117] Step 702: Determine the key points and their corresponding actual quality scores based on the subjective evaluation results. The key points are used to characterize the abrupt change points in the video quality evaluation level.

[0118] Among them, the key points are the points where users' subjective evaluation ratings change abruptly.

[0119] Specifically, key points are used to characterize abrupt changes in video quality ratings, helping the model better understand and simulate key changes in the actual evaluation process. Furthermore, subjective rating results reflect human perceptions of video quality; analyzing these results to identify key points ensures higher model accuracy at critical points of change, thereby improving the overall accuracy of the evaluation.

[0120] Step 704: Based on the first value range and the second value range, continuously iterate and correct the first model parameters and the second model parameters until the deviation between the output value of the corrected video quality objective evaluation model and the actual quality score value of the corresponding key point is not greater than the preset deviation.

[0121] Specifically, when the deviation between the output value of the corrected video quality objective evaluation model and the actual quality score value of the center key point or its vicinity corresponding to the subjective evaluation is not greater than the preset deviation, for example, the preset deviation in this embodiment is 10%, when the deviation ratio between the output value of the corrected video quality objective evaluation model and the actual quality score value corresponding to the subjective evaluation does not exceed 10%, or when the deviation ratio is close to about 10% after two consecutive rounds of optimization, it indicates that the video quality objective evaluation model has been corrected and the result is acceptable.

[0122] For example, under the same playback environment, subjective ratings of video quality by personnel are conducted first to correct the objective video quality evaluation model. The same playback environment refers to the same set-top box, the same network, and the same video source. The video is played for five minutes, and three viewers visually observe and record the network packet loss value when their perception is "very good," "good," "average," "poor," or "very poor." The experiment uses a network impairment device to gradually change the network packet loss rate; initially, there is no packet loss.

[0123] The video quality also starts from a very good state. When two out of the three viewers feel it is no longer "very good," the network impairment change is immediately stopped, and the current network impairment value X1 is recorded. Simultaneously, the average video quality rating Y1 given by the viewers at this point is also recorded. This process is repeated sequentially to obtain (Xi, Yi), where i∈N. * .

[0124] Subjective video quality rating by viewers only needs to be done once. Record the network impairment values ​​X1, X2, X3, X4, and X5, and the average video quality ratings Y1, Y2, Y3, Y4, and Y5 from the viewers.

[0125] By revising and continuously optimizing the model coefficients m and n, we attempt to make the objective evaluation model curve approximate the key points and trends of the subjective score.

[0126] Optimization objective: MIN T=MIN∑[y(Xi)-Yi] 2 =([y(X1)-Y1]) 2 +[y(X2)-Y2] 2 +[y(X3)-Y3] 2 +[y(X4)-Y4] 2 +[y(X5)-Y5] 2 The optimization method involves a hierarchical approximation of coefficient values. The initial value of m is 10, and the initial value of n is 20. In the first round, the step size of m is 1, and the step size of n is 2. All permutations are used, and approximately one hundred optimization calculations are performed to obtain the first round optimization points m1 and n1. Based on m1 and n1, the step size is reduced by a factor of ten. In the second round, the step size of m is 0.1, and the step size of n is 0.2. Another approximately one hundred optimization calculations are performed to obtain the second round optimization points m2 and n2. This patent only uses two rounds of hierarchical approximation calculations in its experiments.

[0127] Finally, with the goal of minimizing the sum of squared differences, a hierarchical approximation optimization algorithm was used to correct the y(x) model result. The corrected y(x) model can be practically applied to video evaluation scenarios. The set-top box probe obtains network packet loss values, which are then substituted into the model calculation to provide a quality evaluation value for the video playback.

[0128] In this embodiment, the model performance is gradually optimized by iteratively correcting the model parameters, making it closer to the optimal solution.

[0129] In a detailed embodiment, as shown in Figure 8, the objective video quality evaluation model includes a set-top box player, a network impairment device control, a network packet loss video objective evaluation model, and a set-top box probe. The set-top box player is used to play the video, and an observer visually observes and makes a subjective evaluation. The set-top box player controls the video playback and displays it on a television set. The network impairment device is connected to the network where the set-top box is located and controls the network packet loss rate of the set-top box video. The observer subjectively evaluates their experience while watching the television video. The network packet loss video objective evaluation model algorithm is integrated into the set-top box probe.

[0130] Determine an objective evaluation model for video quality in the event of network packet loss: y(x) = 5 / [m (n*x)In the experiment, subjective observation of video evaluation was conducted, with five levels (Excellent, Good, Average, Poor, Very Poor). When the video quality level changed abruptly, the network impairment control subjectively observed network impairment critical point (Xi, critical packet loss value) was determined. X1 = 0.001 (one-thousandth), X2 = 0.005 (five-thousandths), X3 = 0.01 (one-hundredth), X4 = 0.02 (two-hundredths), X5 = 0.1 (one-tenth). The subjectively observed critical point video quality score (Yi, subjective score) was: Y1 = 4.8, Y2 = 4, Y3 = 3.2, Y4 = 2.4, Y5 = 1.6. The central key point corresponding to the subjective evaluation was (X3, Y3).

[0131] After two rounds of iterative optimization, the optimized model parameters are shown in the table below:

[0132] According to the table, the optimized model coefficients are: first model parameter m = 6.8, second model parameter n = 17.8. Therefore, the practical application model formula in this embodiment is: y(x) = 5 / [6.8]. (17.8*x) ].

[0133] This embodiment employs the above method, using the user's subjective evaluation level of the target video to iteratively correct the first and second model parameters in the preset objective video quality evaluation model. This makes the output value of the objective video quality evaluation model continuously approach the user's subjective evaluation level, and the corrected objective video quality evaluation model whose final output accuracy meets the preset accuracy threshold is determined as the target video quality evaluation model. This helps to improve the accuracy and reliability of objectively evaluating the quality of the target video.

[0134] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0135] Based on the same inventive concept, this application also provides a video quality evaluation apparatus for implementing the video quality evaluation method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more video quality evaluation apparatus embodiments provided below can be found in the limitations of the video quality evaluation method described above, and will not be repeated here.

[0136] In one embodiment, as shown in FIG9, a video quality evaluation device is provided, including: an evaluation data acquisition module 902, a model parameter optimization module 904, an evaluation model generation module 906, and a video quality evaluation module 908, wherein:

[0137] The evaluation data acquisition module 902 is used to acquire the target video, the preset objective evaluation model for video quality, and the preset subjective evaluation level classification. The first model parameter in the objective evaluation model for video quality is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective evaluation model for video quality is used to characterize the degree of correction of the influence of network packet loss rate on video quality.

[0138] The model parameter optimization module 904 is used to determine the first value range for optimizing the first model parameter according to the subjective evaluation level classification, and to determine the second value range for optimizing the second model parameter according to the subjective evaluation level classification.

[0139] The evaluation model generation module 906 is used to obtain the subjective evaluation level results of multiple users for the target video. Based on the first value range, the second value range and the subjective evaluation level results, iteratively corrects the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thus obtaining the target video quality evaluation model.

[0140] The video quality evaluation module 908 is used to determine the quality score of the target video based on the target video quality evaluation model.

[0141] In one embodiment, the model parameter optimization module 904 is further configured to: obtain a preset mapping table between subjective evaluation level classification and evaluation score; determine the highest evaluation score corresponding to the highest evaluation level and the lowest evaluation score corresponding to the lowest evaluation level; determine the upper limit of the first model parameter based on the highest evaluation score; determine the lower limit of the first model parameter based on the lowest evaluation score; and determine the first value range based on the lower limit and the upper limit of the first model parameter.

[0142] In one embodiment, the model parameter optimization module 904 is further configured to: obtain the maximum packet loss rate of the wide area network where the target video is located; determine the initial value range of the second model parameter based on the maximum packet loss rate; determine the packet loss rate as the independent variable parameter of the video quality objective evaluation model; sample within the initial value range to obtain sampled data, the sampled data including the network packet loss rate value and the evaluation value obtained after inputting the network packet loss rate value into the video quality objective evaluation model; determine the upper limit of the second model parameter based on each evaluation value; and update the initial value range based on the upper limit of the second model parameter to obtain the second value range.

[0143] In one embodiment, the evaluation model generation module 906 is further configured to: determine a first initial value for the first model parameter and a second initial value for the second model based on a first value range and a second value range, respectively; iterate the first initial value and the second initial value a preset number of times with equal step size to obtain multiple first quality score output values; determine updated first initial values ​​and updated second initial values ​​based on the first quality score output values; iterate the updated first initial value and the updated second initial value a preset number of times with equal step size to obtain multiple second quality score output values; and continue to optimize the updated first initial value and the updated second initial value based on the second quality score output values ​​until the output accuracy of the corrected video quality objective evaluation model meets a preset accuracy threshold.

[0144] In one embodiment, the evaluation model generation module 906 is further configured to: determine the corresponding network impairment value based on each first quality score output value; obtain the subjective evaluation value of at least one user based on the network impairment value; and determine the average value of the subjective evaluation value; obtain a preset deviation optimization function; input the one-to-one corresponding network impairment value and the average value of the subjective evaluation value into the deviation optimization function to obtain the deviation value; determine the first initial value corresponding to the minimum deviation value as the updated first initial value; and determine the second initial value corresponding to the minimum variation value as the updated second initial value.

[0145] In one embodiment, the evaluation model generation module 906 is further configured to: determine key points and their corresponding actual quality scores based on the subjective evaluation level results, wherein the key points are used to characterize the abrupt change points of the video quality evaluation level; and iteratively correct the first model parameters and the second model parameters based on the first value range and the second value range until the deviation between the output value of the corrected video quality objective evaluation model and the actual quality score of the corresponding key point is not greater than a preset deviation.

[0146] Each module in the aforementioned video quality evaluation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0147] In one embodiment, a computer device, which may be a server, is provided, and its internal structure is shown in Figure 10. The computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores user subjective evaluation data for a target video, corresponding network packet loss rates, and other data. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a video quality evaluation method.

[0148] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in Figure 11. The computer device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a video quality evaluation method. The display screen of the computer device may be a liquid crystal display (LCD) or an e-ink display. The input device of the computer device may be a touch layer covering the display screen, or buttons, a trackball, or a touchpad located on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0149] Those skilled in the art will understand that the structures shown in Figures 10 and 11 are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements.

[0150] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0151] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0152] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0153] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0154] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0155] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0156] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A video quality evaluation method, characterized in that, The method includes: The system acquires a target video, a preset objective video quality evaluation model, and a preset subjective evaluation level classification. The first model parameter in the objective video quality evaluation model is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective video quality evaluation model is used to characterize the degree of correction of the influence of network packet loss rate on video quality. Based on the subjective evaluation level classification, a first value range for optimizing the first model parameters is determined, and based on the subjective evaluation level classification, a second value range for optimizing the second model parameters is determined. Obtain subjective evaluation results from multiple users for the target video. Based on the first value range, the second value range, and the subjective evaluation results, continuously iterate and correct the first model parameters and the second model parameters until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thereby obtaining the target video quality evaluation model. The quality score of the target video is determined based on the target video quality evaluation model.

2. The method according to claim 1, characterized in that, The step of determining the first value range for optimizing the first model parameters based on the subjective evaluation level classification includes: Obtain the preset mapping table between subjective evaluation level categories and evaluation scores, and determine the highest evaluation score corresponding to the highest evaluation level and the lowest evaluation score corresponding to the lowest evaluation level. Based on the highest evaluation score, the upper limit of the first model parameter is determined, and based on the lowest evaluation score, the lower limit of the first model parameter is determined. The first value range is determined based on the lower limit and the upper limit of the first model parameter.

3. The method according to claim 1, characterized in that, The step of determining the second value range for optimizing the second model parameters based on the subjective evaluation level classification includes: Obtain the maximum packet loss rate of the wide area network where the target video is located, and determine the initial value range of the second model parameters based on the maximum packet loss rate; The transmission packet loss rate is determined as the independent variable parameter of the objective video quality evaluation model. Sampling is performed within the initial value range to obtain sampled data. The sampled data includes the network packet loss rate value and the evaluation value obtained after inputting the network packet loss rate value into the objective video quality evaluation model. Based on each of the evaluation values, the upper limit of the second model parameter is determined, and based on the upper limit of the second model parameter, the initial value range is updated to obtain the second value range.

4. The method according to claim 1, characterized in that, The step of iteratively correcting the first model parameters and the second model parameters based on the first value range and the second value range until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold includes: Based on the first value range and the second value range, the first initial value of the first model parameter and the second initial value of the second model are determined respectively; The first initial value and the second initial value are iterated for a preset number of times with equal step size to obtain multiple first quality score output values. Based on the first quality score output values, the updated first initial value and the updated second initial value are determined. The updated first initial value and the updated second initial value are iterated a preset number of times with equal step size to obtain multiple second quality score output values. Based on the second quality score output values, the updated first initial value and the updated second initial value are further optimized until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold.

5. The method according to claim 4, characterized in that, The step of determining the updated first initial value and the updated second initial value based on the first quality score output value includes: Based on each of the first quality score output values, the corresponding network impairment value is determined. Based on the network impairment value, the subjective evaluation value of at least one user is obtained, and the average value of the subjective evaluation value is determined. Obtain a preset deviation optimization function, input the average value of the network impairment value and the subjective evaluation value that correspond one-to-one into the deviation optimization function to obtain the deviation value, determine the first initial value corresponding to the minimum deviation value as the updated first initial value, and determine the second initial value corresponding to the minimum variation value as the updated second initial value.

6. The method according to claim 1, characterized in that, The step of iteratively correcting the first model parameters and the second model parameters based on the first value range, the second value range, and the subjective evaluation level result until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold includes: Based on the subjective evaluation level results, key points and their corresponding actual quality scores are determined, and the key points are used to characterize the abrupt change points in the video quality evaluation level. Based on the first value range and the second value range, the first model parameters and the second model parameters are iteratively corrected until the deviation between the output value of the corrected video quality objective evaluation model and the actual quality score value of the corresponding key point is not greater than the preset deviation.

7. A video quality evaluation device, characterized in that, The device includes: The evaluation data acquisition module is used to acquire the target video, the preset objective evaluation model for video quality, and the preset subjective evaluation level classification. The first model parameter in the objective evaluation model for video quality is used to characterize the degree of influence of network packet loss rate on video quality, and the first model parameter in the objective evaluation model for video quality is used to characterize the degree of correction of the influence of network packet loss rate on video quality. The model parameter optimization module is used to determine a first value range for optimizing the first model parameter based on the subjective evaluation level classification, and to determine a second value range for optimizing the second model parameter based on the subjective evaluation level classification. The evaluation model generation module is used to obtain the subjective evaluation level results of multiple users for the target video, and continuously iterate and correct the first model parameters and the second model parameters according to the first value range, the second value range and the subjective evaluation level results until the output accuracy of the corrected video quality objective evaluation model meets the preset accuracy threshold, thereby obtaining the target video quality evaluation model. The video quality evaluation module is used to determine the quality score of the target video based on the target video quality evaluation model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images