Video enhancement method, apparatus and device, storage medium, and program product

Abstract

Embodiments of the present application provide a video enhancement method, apparatus and device, a storage medium, and a program product. The method comprises: when a video is subjected to enhancement processing, obtaining the watching duration of the video, wherein the enhancement processing corresponds to an original enhancement intensity; determining a corresponding video enhancement coefficient on the basis of the watching duration; and performing enhancement processing on the video on the basis of the video enhancement coefficient and the original enhancement intensity, wherein the longer the watching duration is, the smaller the enhancement processing intensity is. According to the scheme, the duration during which a user watches the video subjected to enhancement processing can be obtained by obtaining the watching duration of the video, the video enhancement coefficient is reasonably adapted to long-term eye habits of the user, and the original enhancement intensity is attenuated and adjusted, such that the long-term subjective experience reduction is effectively relieved while meeting the short-term subjective experience improvement of audiences, the gain of video enhancement on user subjective experience is improved, and the effect of balancing the short-term subjective experience improvement and the long-term subjective experience reduction of audiences is achieved.

Description

Video enhancement method, device, equipment, storage medium and program product

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on December 14, 2023, with application number 202311733785.9, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present application relate to the field of computer technology, and in particular to a video enhancement method, apparatus, device, storage medium, and program product. Background Art

[0003] To improve the display quality of a video, the video can be processed by video enhancement. For example, by enhancing the color of the video, the color perception of the video can be changed to make the video more attractive to users.

[0004] In related technologies, R&D teams are constantly optimizing and improving video enhancement to increase the subjective benefits of video, using more complex algorithms to achieve more effective results. However, in actual applications, it has been found that greater video enhancement does not necessarily lead to longer user viewing time. Actual user viewing time, and therefore user satisfaction, does not match the improved video enhancement results, requiring improvement. Summary of the Invention

[0005] The embodiments of the present application provide a video enhancement method, apparatus, device, storage medium, and program product, which solve the problem of mismatch between the improvement of video enhancement effect and the user's viewing time during the video enhancement process. This solution is more in line with the user's eye habits. While satisfying the audience's short-term subjective experience improvement, it adjusts the original enhancement intensity by attenuation based on the user's viewing time, effectively alleviating the long-term subjective experience degradation, and improving the benefits of video enhancement for the user's subjective feelings, achieving a balance between the audience's short-term subjective experience improvement and the long-term subjective experience degradation.

[0006] In a first aspect, an embodiment of the present application provides a video enhancement method, the method comprising:

[0007] Obtaining a viewing time of the video when enhancement processing is performed on the video, wherein the enhancement processing corresponds to an original enhancement strength;

[0008] Determining a corresponding video enhancement coefficient according to the viewing time;

[0009] The video is enhanced based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time is, the smaller the strength of the enhancement is.

[0010] In a second aspect, an embodiment of the present application further provides a video enhancement device, the device comprising:

[0011] a viewing time acquisition module configured to acquire the viewing time of the video when the video is enhanced, wherein the enhancement processing corresponds to an original enhancement strength;

[0012] an enhancement coefficient determination module, configured to determine a corresponding video enhancement coefficient according to the viewing time;

[0013] The video enhancement processing module is configured to perform enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time is, the smaller the strength of the enhancement processing is.

[0014] In a third aspect, an embodiment of the present application further provides a video enhancement device, the device comprising:

[0015] one or more processors;

[0016] a storage device configured to store one or more programs,

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement the video enhancement method described in the embodiment of the present application.

[0018] In a fourth aspect, an embodiment of the present application further provides a non-volatile storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the video enhancement method described in the embodiment of the present application when executed by a computer processor.

[0019] In a fifth aspect, an embodiment of the present application further provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor of the device reads and executes the computer program from the computer-readable storage medium, so that the device performs the video enhancement method described in the embodiment of the present application.

[0020] In an embodiment of the present application, by obtaining the viewing time of the video while performing enhancement processing on the video, wherein the enhancement processing corresponds to the original enhancement strength; determining the corresponding video enhancement coefficient according to the viewing time; and performing enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the intensity of the enhancement processing. In the above scheme, by obtaining the viewing time of the video, the duration of the user watching the enhanced video can be obtained, the video enhancement coefficient can be determined reasonably to adapt to the user's long-term eye habits, and the original enhancement strength can be attenuated and adjusted. While satisfying the audience's short-term subjective experience improvement, it effectively alleviates the long-term subjective experience reduction, improves the benefits of video enhancement on the user's subjective feelings, and achieves a balance between the audience's short-term subjective experience improvement and the long-term subjective experience reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG1 is a flow chart of a video enhancement method provided in an embodiment of the present application;

[0022] FIG2 is a flow chart of a method for enhancing a video based on a video enhancement coefficient and an original enhancement strength provided by an embodiment of the present application;

[0023] FIG3 is a flow chart of a video enhancement method including a video enhancement coefficient determination process provided by an embodiment of the present application;

[0024] FIG4 is a flow chart of a method for determining a mapping relationship between viewing time and video enhancement coefficient provided by an embodiment of the present application;

[0025] FIG5 is a graph showing average viewing time and AB revenue obtained through an AB experiment according to an embodiment of the present application;

[0026] FIG6 is a flow chart of a method for determining AB revenue under different average viewing times by setting up an AB experiment, provided in an embodiment of the present application;

[0027] FIG7 is a structural block diagram of a video enhancement device provided in an embodiment of the present application;

[0028] FIG8 is a schematic structural diagram of a video enhancement device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0029] The following is a further detailed description of the embodiments of the present application in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the embodiments of the present application, and are not intended to limit the embodiments of the present application. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions of the embodiments of the present application, rather than all structures.

[0030] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0031] The video enhancement method provided in the embodiments of this application can be applied to video output scenarios that improve the display effects of video color, video brightness, etc., such as online video, audio and video calls, video conferencing, and live video broadcasts. Taking the application scenario of online video as an example, the video enhancement intensity can be reasonably adjusted to attenuate as the user's online viewing time increases. The several application scenarios listed above are merely exemplary and explanatory. In actual applications, the video enhancement method can also be used in video output in other scenarios, and the embodiments of this application are not limited to this.

[0032] In the related art, in order to improve the display effect of the video, the video can be processed by video enhancement. For example, by enhancing the color of the video, the color perception of the video can be changed to make the presented video more attractive to users. In a short period of time, video enhancement can bring a visual impact of clarity and color, which is easier to capture the audience's attention and meet the audience's need for improved short-term subjective experience. However, since large-scale color enhancement will cause certain stimulation to the vision, long-term viewing can easily lead to eye fatigue, cause symptoms such as eye soreness and headache, and may even accelerate the aging of the retina and optic nerve, causing hidden damage to vision, thereby reducing the long-term subjective experience. Reducing the benefits of video enhancement for the user's subjective feelings will reduce the user's viewing time. The user's actual viewing time, that is, the user's satisfaction, does not match the improvement of the video enhancement effect. Therefore, the present application aims to provide a video enhancement method to solve the problem of mismatch between the improvement of the video enhancement effect and the user's viewing time during the video enhancement process, so as to achieve a balance between the improvement of the audience's short-term subjective experience and the decline of the long-term subjective experience.

[0033] In the video enhancement method provided in the embodiment of the present application, the execution entity of each step can be a computer device, which refers to any electronic device with data calculation, processing and storage capabilities, such as mobile phones, PCs (Personal Computers), tablet computers and other terminal devices, or servers and other devices. The embodiment of the present application does not limit this.

[0034] FIG1 is a flow chart of a video enhancement method provided by an embodiment of the present application. As shown in FIG1 , the method includes the following steps:

[0035] Step S101: Obtain the viewing time of the video when the video is enhanced, wherein the enhancement process corresponds to an original enhancement strength.

[0036] Enhancement processing can be based on a set video enhancement intensity to enhance the video's visual effects, such as color enhancement, contrast enhancement, and brightness enhancement. This enhancement can enhance the video's colors and visual effects, thereby attracting users. The original enhancement intensity is the original enhancement intensity used during video enhancement. For example, to enhance the color saturation of a video, this can be the intensity used in the enhancement algorithm to increase saturation. A higher intensity increases the enhancement effect, while a lower intensity decreases the effect. Optionally, the size of the original enhancement intensity can be adaptively set by the developer based on the enhancement algorithm and the content being enhanced. The specific value can be calculated using parameters fitted from training data and deployed in a computing device, such as a server, to enhance the input video. The original enhancement intensity can be adaptively determined based on the identified features after feature recognition of the input video. For example, if brightness enhancement is required, the intensity of the brightness enhancement can be adaptively determined, representing the original enhancement intensity.

[0037] In one embodiment, when the video is enhanced, the viewing time of the video is obtained. Optionally, the viewing time of the video can be determined based on different methods. For example, the cumulative time from the start time node of the video to the current time node can be used as the viewing time of the video. Among them, the video playback time as the viewing time may deviate from the user's actual viewing time. It may happen that the user's sight temporarily leaves the video screen while the video is still playing. Optionally, the image of the user watching the video can be captured by the image acquisition device of the user terminal, such as a camera, to confirm the user's actual viewing time, so as to achieve the purpose of accurately calculating the viewing time of the video, but it may increase the occupation of computing resources and the complexity. Of course, there is no specific limitation on how to determine the viewing time of the video, and any method that can reflect the user's cumulative viewing time is applicable to this embodiment.

[0038] In one embodiment, the video viewing time can be obtained periodically based on a preset time interval, for example, once every n milliseconds, and the specific value of n can be set based on the network conditions and computing performance resources of the actual application scenario. In addition, it can also be obtained based on a pre-set non-fixed time interval duration node sequence, for example, the duration node sequence can be set to n1, n2, ..., n i , where n i Can indicate continuous n i The duration node of time, n i The unit can be milliseconds, seconds, minutes, etc. Of course, other timing acquisition methods can also be used, and this application does not limit this.

[0039] Step S102: Determine a corresponding video enhancement coefficient according to the viewing time.

[0040] Among them, since the purpose of video enhancement is to allow users to obtain a better subjective experience, increase the user's viewing time, and improve the time benefit, under the enhanced effect of the original enhancement intensity, the demand for improving the user's subjective experience in a short time can be achieved. However, as the viewing time increases, the continuous enhancement output violates the user's eye habits and will instead lead to a decrease in the time benefit. Therefore, this embodiment corresponds to setting a video enhancement coefficient, which can be an attenuation proportional coefficient for adjusting the original enhancement intensity based on the length of the viewing time. For example, the original enhancement intensity can be attenuated and adjusted as the viewing time increases, reducing the video enhancement effect to alleviate the problem of reduced long-term subjective experience of users.

[0041] In one embodiment, the correspondence between viewing time and video enhancement coefficient can be determined based on different data learning methods. For example, in order to alleviate the long-term reduction in user subjective experience, as the viewing time increases, the video enhancement strength can be attenuated to different degrees based on the video enhancement coefficient. Thus, the commonly used attenuation function relationship can be directly used as the correspondence between viewing time and video enhancement coefficient. For example, exponential attenuation function, logarithmic attenuation function, polynomial attenuation function, etc., by selecting different attenuation functions to simulate the process of video enhancement strength attenuating with increasing viewing time at different attenuation degrees. Of course, the actual selection of the attenuation function and the coefficient adjustment can be determined based on the revenue data feedback after application, which is not limited in this application.

[0042] Optionally, it is also possible to set up an online experiment. For example, an AB experiment can be set up to count the viewing time of videos with and without video enhancement, and fit the correspondence between viewing time and video enhancement coefficient from the relationship between online viewing time and time benefit. For another example, multiple parallel control groups corresponding to different preset correspondences can be set up. The preset correspondences can be different attenuation function relationships or different coefficient combinations of the same attenuation function. The correspondence between viewing time and video enhancement coefficient with the highest matching degree is determined by comparing the time benefits of the parallel control groups. Of course, the determination of the correspondence between viewing time and video enhancement coefficient is an exemplary explanation, and the correspondence that can reflect the attenuation of video enhancement strength with the increase of viewing time is within the scope of protection of this application.

[0043] Step S103: performing enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the enhancement strength.

[0044] After determining the video enhancement coefficient based on the current video viewing time, the video is enhanced by combining the video enhancement coefficient with the original enhancement strength. It is understandable that as viewing time increases, maintaining the original enhancement strength while enhancing the video will result in a decrease in the long-term subjective experience. Therefore, the video enhancement coefficient can be adjusted to attenuate the original enhancement strength as viewing time increases. This results in a reduction in the enhancement strength with longer viewing time.

[0045] In one embodiment, the original enhancement strength is reduced according to a preset calculation rule based on the video enhancement coefficient to obtain a target enhancement strength. For example, the video enhancement coefficient can be a decay ratio coefficient corresponding to the viewing time of the current video. FIG2 is a flowchart of a method for enhancing a video based on the video enhancement coefficient and the original enhancement strength provided in an embodiment of the present application. As shown in FIG2, the method specifically includes the following steps:

[0046] Step S1031: Multiply the video enhancement coefficient by the original enhancement strength to obtain the target enhancement strength.

[0047] Step S1032: Enhance the video based on the target enhancement strength.

[0048] Among them, by multiplying the video enhancement coefficient with the original enhancement strength, the original enhancement strength is reduced according to the preset attenuation ratio to obtain the target enhancement strength, which can effectively adapt to the increase in viewing time and weaken the enhancement processing, which is more in line with the user's long-term viewing needs and improves the viewing benefits of long-term users.

[0049] In one embodiment, the obtained video enhancement coefficient can also be fine-tuned according to the intensity level of the original enhancement intensity. For the original enhancement intensity at a high intensity level, the video enhancement coefficient can be lowered by a preset amount to increase the attenuation rate, which is conducive to quickly alleviating the visual fatigue caused by a short-term and substantial increase in subjective experience; for the original enhancement intensity at a low intensity level, the video enhancement coefficient can be raised by a preset amount to reduce the attenuation rate, which is conducive to maintaining a good visual experience caused by a short-term and slight increase in subjective experience. Of course, the above calculation of the target enhancement intensity by the video enhancement coefficient and the original enhancement intensity is an exemplary explanation, and the adaptive change settings of other calculation rules adapted to different application scenarios are all within the scope of protection of this application.

[0050] From the above, it can be seen that by obtaining the viewing time of the video while performing enhancement processing on the video, wherein the enhancement processing corresponds to the original enhancement strength; determining the corresponding video enhancement coefficient according to the viewing time; and performing enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the intensity of the enhancement processing. In the above scheme, by obtaining the viewing time of the video, the duration of the user watching the enhanced video can be obtained, and the video enhancement coefficient is determined to reasonably adapt to the user's long-term eye habits, and the original enhancement strength is attenuated and adjusted. While satisfying the audience's short-term subjective experience improvement, it effectively alleviates the long-term subjective experience reduction, improves the benefits of video enhancement on the user's subjective feelings, and achieves a balance between the audience's short-term subjective experience improvement and the long-term subjective experience reduction.

[0051] FIG3 is a flow chart of a video enhancement method including a video enhancement coefficient determination process provided by an embodiment of the present application. As shown in FIG3 , the method includes:

[0052] Step S201: Obtain the viewing time of the video when the video is enhanced, wherein the enhancement processing corresponds to the original enhancement strength.

[0053] Step S202: Determine the video enhancement coefficient corresponding to the viewing time according to the viewing time and the set mapping relationship between the viewing time and the video enhancement coefficient.

[0054] Among them, the mapping relationship can be a numerical correspondence between viewing time and video enhancement coefficient, and each viewing time obtained can be matched to the corresponding video enhancement coefficient through the mapping relationship. The mapping relationship can include an attenuation function relationship, for example, it can be a commonly used attenuation function relationship, such as an exponential attenuation function, a logarithmic attenuation function, a polynomial attenuation function, etc., and for example, it can be a function relationship obtained by data fitting based on online test data; it can also include a mapping table, for example, multiple time ranges are pre-set, and each time range corresponds to a video enhancement coefficient. The time range in which the current viewing time is located is determined by querying the mapping table, and then the corresponding video enhancement coefficient is determined. Of course, there can also be other settings for the mapping relationship between viewing time and video enhancement coefficient, which are not limited in this application.

[0055] Step S203: Enhance the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the enhancement strength.

[0056] From the above, we can see that by establishing a mapping relationship between viewing time and video enhancement coefficient and accurately fitting the corresponding relationship between viewing time and video enhancement coefficient, we can match reasonable video enhancement coefficients for different viewing times, and then effectively adapt to changes in user viewing time to reasonably adjust the video enhancement intensity and ensure the benefits of long-time video.

[0057] FIG4 is a flow chart of a method for determining a mapping relationship between viewing time and video enhancement coefficient provided by an embodiment of the present application, as shown in FIG4 , including:

[0058] Step S301: Determine AB revenues under different average viewing times through a set AB experiment, wherein the AB experiment includes statistics on viewing times of videos with and without video enhancement.

[0059] The AB experiment may be performed by using videos with and without video enhancement as the experimental and control groups, and determining AB benefits for different average viewing time ranges. The AB benefits may reflect the difference in user viewing time for videos with and without video enhancement. For example, the AB benefits may be calculated as the difference between the average viewing time of videos with and without video enhancement within the same average viewing time range. Taking the statistical data of an AB experiment as an example, Table 1 shows the recorded data of different average viewing times and corresponding AB benefits. Table 1 is as follows:

[0060] Table 1

[0061] It can be seen from Table 1 that when the average viewing time is 0-5s, the AB benefit is 0.7s, when the average viewing time is 5-10s, the AB benefit is 0.3s, when the average viewing time is 10-15s, the AB benefit is 0.1s, when the average viewing time is 15-20s, the AB benefit is 0.007s, when the average viewing time is 20-25s, the AB benefit is -0.1s, when the average viewing time is 25-30s, the AB benefit is -0.13s, and when the average viewing time is 30-35s, the AB benefit is -0.25s. Based on the statistical data, Figure 5 can be obtained. Figure 5 is a curve diagram of the average viewing time and AB benefit obtained through an AB experiment provided in an embodiment of the present application. As shown in Figure 5, as the average viewing time increases, the AB benefit brought by video enhancement gradually decreases, and even reaches a negative value. Therefore, by setting up an AB experiment to determine the AB benefits of different average viewing times, it is possible to effectively reflect the impact of video enhancement intensity on the user viewing experience as the viewing time increases.

[0062] Optionally, FIG6 is a flowchart of a method for determining AB revenue under different average viewing times by setting up an AB experiment provided in an embodiment of the present application. As shown in FIG6 , the method includes the following steps:

[0063] Step S3011: Filter the anchors with positive video enhancement effects from the results of the set AB experiment.

[0064] Among them, the positive effect of video enhancement can be determined based on the subjective scoring of the anchor's video by technical personnel after a short period of viewing, excluding anchors with negative video enhancement effects, to ensure that the video enhancement in the result data of the AB experiment can achieve an improvement in the short-term user subjective experience.

[0065] Step S3012: Sort the selected anchors according to their corresponding average viewing time.

[0066] Among them, sorting the anchors according to their corresponding average viewing time can provide a range reference for different average viewing times, which is conducive to the subsequent sorting of AB benefits corresponding to different average viewing times.

[0067] Step S3013: When the average viewing time used in sorting is the enhanced average viewing time after video enhancement, obtain the corresponding non-enhanced average viewing time without video enhancement.

[0068] Among them, in the set AB experiment, the enhanced average viewing time after video enhancement and the unenhanced average viewing time without video enhancement can be used as comparative data, reflecting the difference in users' viewing perception of video enhancement and video without video enhancement within the same average viewing time range. Therefore, there will be numerical differences in the enhanced average viewing time and the unenhanced average viewing time, which can reflect the AB benefits of the AB experiment. For example, when the enhanced average viewing time is greater than the unenhanced average viewing time, the AB benefit is positive; when the enhanced average viewing time is less than the unenhanced average viewing time, the AB benefit is negative.

[0069] Step S3014: Calculate AB benefits based on the enhanced average viewing time and the non-enhanced average viewing time.

[0070] Among them, the AB benefit can be obtained by calculating the difference or ratio of the enhanced average viewing time and the non-enhanced average viewing time, which reflects the degree of improvement of the user's viewing experience due to video enhancement as the viewing time increases.

[0071] By combining the average duration data of the host video in different video enhancement states, we can effectively determine the AB benefits under different average viewing times, accurately reflect the changing patterns of users' long-term viewing experience after video enhancement, and improve data reliability.

[0072] Step S302: fitting is performed based on different average viewing durations and corresponding AB benefits to obtain a duration-benefit distribution.

[0073] Among them, since AB benefits can reflect that users' acceptance of video enhancement effects decreases with the increase of average viewing time, the video enhancement strength should be attenuated to adapt to the decline of AB benefits. The duration benefit distribution can accurately reflect the attenuation law of AB benefits as the average viewing time increases, and then can characterize the attenuation law of video enhancement strength as the average viewing time increases. Therefore, we can first fit different average viewing times and corresponding AB benefits to obtain the duration benefit distribution, and reasonably deduce the relationship between video enhancement coefficient and duration based on the relationship between online viewing time and duration benefit, providing a reliable reference relationship for the correspondence between video enhancement coefficient and viewing time, thereby improving credibility.

[0074] Optionally, the duration benefit distribution can be: y1=-Aln(x1)+B,

[0075] Here, y1 represents the AB revenue, x1 represents the average viewing time, A represents the first undetermined non-negative coefficient, and B represents the second undetermined coefficient. The duration revenue distribution exhibits a logarithmic decay pattern, effectively fitting the mapping relationship between AB revenue and average viewing time.

[0076] Step S303: Based on the duration benefit distribution and the normalized enhancement coefficient data and duration data, a mapping relationship between viewing time and video enhancement coefficient is calculated, wherein the end values ​​of the enhancement coefficient data correspond to the maximum enhancement intensity and the minimum enhancement intensity of the enhancement processing, and the end values ​​of the duration data correspond to the average viewing time when AB benefits are non-negative maximum and non-negative minimum values.

[0077] The enhancement coefficient data may be statistical data showing the decay of video enhancement intensity over time. The enhancement coefficient data may be obtained by normalizing the video enhancement intensity corresponding to the video in the video enhancement state. The first end value of the enhancement coefficient data corresponds to the maximum enhancement intensity, representing the strongest enhancement effect, and the second end value corresponds to the minimum enhancement intensity, indicating that no enhancement is required and the original video remains unchanged. For example, if the video enhancement intensity is normalized to [0, 1], then 0 indicates no enhancement is required and 1 indicates the strongest enhancement effect. The duration data may be statistical data on video viewing time or length. The duration data may be obtained by normalizing the average viewing time of the video in the video enhancement state. The first end value of the duration data corresponds to the average viewing time when the AB gain is a non-negative maximum value, representing the time when the video enhancement is at its maximum intensity. The second end value corresponds to the average viewing time when the AB gain is a non-negative minimum value, representing the time when the video enhancement needs to be stopped. For example, if the average viewing time is normalized to [0, 1], then 0 indicates the time when the video enhancement is at its maximum intensity and 1 indicates the time when the video enhancement needs to be stopped. By substituting the normalized enhancement coefficient data and duration data into the functional relationship corresponding to the duration-revenue distribution, the correlation coefficient of the functional relationship can be obtained, thereby completing the fitting of the mapping relationship between viewing time and video enhancement coefficient.

[0078] Optionally, the mapping relationship can be: y2=-λln(x2)+∈,

[0079] Among them, y2 represents the video enhancement coefficient, x2 represents the viewing time, λ represents the first non-negative coefficient, and ∈ represents the second coefficient. This mapping relationship is a logarithmic attenuation function relationship, which effectively associates the video enhancement coefficient with the average viewing time, accurately reflecting the changing law of the video enhancement coefficient attenuating with the increase of time length.

[0080] From the above, we can see that by conducting AB benefit analysis on the set AB experiment, we can effectively determine the regular distribution of AB benefits and average viewing time, and then reasonably deduce the mapping relationship between the video enhancement coefficient and the average viewing time. This can provide a reasonable attenuation ratio of video enhancement intensity for different viewing times, thereby achieving the goal of balancing users' short-term subjective feelings and long-term subjective feelings.

[0081] FIG7 is a block diagram of a video enhancement device provided in an embodiment of the present application. The device is configured to execute the video enhancement method provided in the above embodiment and has the corresponding functional modules and beneficial effects of the execution method. As shown in FIG7 , the device specifically includes:

[0082] The viewing time acquisition module 101 is configured to acquire the viewing time of the video when the video is enhanced, wherein the enhancement processing corresponds to the original enhancement strength;

[0083] An enhancement coefficient determination module 102 is configured to determine a corresponding video enhancement coefficient according to the viewing time;

[0084] The video enhancement processing module 103 is configured to perform enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the enhancement strength.

[0085] In the above, by obtaining the viewing time of the video while performing enhancement processing on the video, wherein the enhancement processing corresponds to the original enhancement strength; determining the corresponding video enhancement coefficient according to the viewing time; and performing enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the intensity of the enhancement processing. In the above scheme, by obtaining the viewing time of the video, the duration of the user watching the enhanced video can be obtained, the video enhancement coefficient can be determined reasonably to adapt to the user's long-term eye habits, and the original enhancement strength can be attenuated and adjusted. While satisfying the audience's short-term subjective experience improvement, the long-term subjective experience reduction can be effectively alleviated, and the benefits of video enhancement on the user's subjective feelings can be increased, achieving a balance between the audience's short-term subjective experience improvement and the long-term subjective experience reduction.

[0086] In a possible embodiment, the enhancement coefficient determination module 102 is configured to:

[0087] The video enhancement coefficient corresponding to the viewing time is determined according to the viewing time and the set mapping relationship between the viewing time and the video enhancement coefficient.

[0088] In a possible embodiment, the enhancement coefficient determination module 102 is configured to:

[0089] Determine the AB revenue under different average viewing times through the set-up AB experiment. The AB experiment includes statistics on the viewing time of videos with and without video enhancement.

[0090] The duration-benefit distribution is obtained by fitting different average viewing times and corresponding AB benefits;

[0091] The mapping relationship between viewing time and video enhancement coefficient is calculated based on the duration benefit distribution and the normalized enhancement coefficient data and duration data, wherein the end values ​​of the enhancement coefficient data correspond to the maximum enhancement intensity and the minimum enhancement intensity of the enhancement processing, and the end values ​​of the duration data correspond to the average viewing time when AB benefits are non-negative maximum and non-negative minimum values, respectively.

[0092] In a possible embodiment, the enhancement coefficient determination module 102 is further configured to:

[0093] Filter the anchors with positive video enhancement effects from the results of the set AB experiment;

[0094] Sort the selected anchors according to their corresponding average viewing time;

[0095] When the average viewing time used in sorting is the enhanced average viewing time after video enhancement, obtain the corresponding non-enhanced average viewing time without video enhancement;

[0096] AB benefits are calculated based on the enhanced average viewing time and the non-enhanced average viewing time.

[0097] In a possible embodiment, the duration benefit distribution includes: y1 = -Aln(x1) + B,

[0098] Among them, y1 represents AB revenue, x1 represents the average viewing time, A represents the first non-negative coefficient to be determined, and B represents the second coefficient to be determined.

[0099] In a possible embodiment, the mapping relationship includes: y2=-λln(x2)+∈,

[0100] Where y2 represents the video enhancement coefficient, x2 represents the viewing time, λ represents the first non-negative coefficient, and ∈ represents the second coefficient.

[0101] In a possible embodiment, the video enhancement processing module 103 is configured as follows:

[0102] Multiply the video enhancement coefficient by the original enhancement strength to obtain the target enhancement strength;

[0103] Enhance the video based on the target enhancement strength.

[0104] FIG8 is a schematic diagram of the structure of a video enhancement device provided in an embodiment of the present application. As shown in FIG8 , the device includes a processor 201, a memory 202, an input device 203, and an output device 204. The device may include one or more processors 201, with FIG8 assuming a single processor 201. The processor 201, memory 202, input device 203, and output device 204 may be connected via a bus or other means, with FIG8 assuming a bus connection as an example. Memory 202, as a computer-readable storage medium, can be configured to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the video enhancement method in the embodiment of the present application. Processor 201 executes the software programs, instructions, and modules stored in memory 202 to execute various functional applications and data processing of the device, thereby implementing the aforementioned video enhancement method. Input device 203 can be configured to receive input digital or character information and generate key signal input related to user settings and function control of the device. Output device 204 may include a display device such as a display screen.

[0105] An embodiment of the present application also provides a non-volatile storage medium containing computer-executable instructions, which, when executed by a computer processor, is configured to execute a video enhancement method described in the above embodiment, which includes: obtaining the viewing time of the video when enhancing the video, wherein the enhancement processing corresponds to an original enhancement strength; determining a corresponding video enhancement coefficient based on the viewing time; and enhancing the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time, the smaller the strength of the enhancement processing.

[0106] It is worth noting that in the embodiment of the above-mentioned video enhancement device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not configured to limit the scope of protection of the embodiments of this application.

[0107] In some possible implementations, various aspects of the methods provided herein may also be implemented in the form of a program product, which includes program code. When the program product is executed on a computer device, the program code is configured to cause the computer device to perform the steps of the methods according to the various exemplary embodiments of the present application described above. For example, the computer device may perform the video enhancement method described in the embodiments of the present application. The program product may be implemented using any combination of one or more readable media.

Claims

1. A video enhancement method, wherein: include: In the case of performing enhancement processing on the video, obtaining the viewing time of the video, wherein the enhancement processing corresponds to the original enhancement strength; Determining a corresponding video enhancement coefficient according to the viewing time; The video is enhanced based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time is, the smaller the strength of the enhancement is.

2. The video enhancement method according to claim 1, wherein: The determining of the corresponding video enhancement coefficient according to the viewing time includes: The video enhancement coefficient corresponding to the viewing time is determined according to the viewing time and the mapping relationship between the set viewing time and the video enhancement coefficient.

3. The video enhancement method according to claim 2, wherein: The process of determining the mapping relationship includes: Determine AB benefits under different average viewing durations by setting up AB experiments, wherein the AB experiments include statistics on viewing durations of videos with and without video enhancement; Based on the different average viewing times and the corresponding AB benefits, fitting is performed to obtain a time benefit distribution; The mapping relationship between viewing time and video enhancement coefficient is calculated based on the duration benefit distribution and the normalized enhancement coefficient data and duration data. The end values ​​of the enhancement coefficient data correspond to the maximum enhancement intensity and the minimum enhancement intensity of the enhancement processing, and the end values ​​of the duration data correspond to the average viewing time when AB benefits are non-negative maximum and non-negative minimum values, respectively.

4. The video enhancement method according to claim 3, wherein: The AB experiment is set up to determine the AB benefits under different average viewing times, including: Select anchors with positive video enhancement effects from the results of the set AB experiment; Sort the selected anchors according to their corresponding average viewing time; When the average viewing time used in sorting is the enhanced average viewing time after video enhancement, obtaining the corresponding non-enhanced average viewing time without video enhancement; The AB benefit is calculated based on the enhanced average viewing time and the non-enhanced average viewing time.

5. The video enhancement method according to claim 3, wherein: The duration benefit distribution includes: y1 = -Aln(x1) + B, Among them, y1 represents AB revenue, x1 represents the average viewing time, A represents the first non-negative coefficient to be determined, and B represents the second coefficient to be determined.

6. The video enhancement method according to any one of claims 2 or 3, wherein: The mapping relationship includes: y2=-λln(x2)+ε, Wherein, y2 represents the video enhancement coefficient, x2 represents the viewing time, λ represents the first non-negative coefficient, and ε represents the second coefficient.

7. The video enhancement method according to any one of claims 1 to 6, wherein: The step of performing enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength includes: Multiplying the video enhancement coefficient by the original enhancement strength to obtain a target enhancement strength; The video is enhanced based on the target enhancement strength.

8. A video enhancement device, wherein: include: A viewing time acquisition module, configured to acquire the viewing time of the video when the video is enhanced, wherein the enhancement processing corresponds to an original enhancement strength; An enhancement coefficient determination module, configured to determine a corresponding video enhancement coefficient according to the viewing time; The video enhancement processing module is configured to perform enhancement processing on the video based on the video enhancement coefficient and the original enhancement strength, wherein the longer the viewing time is, the smaller the strength of the enhancement processing is.

9. A video enhancement device, comprising: one or more processors; A storage device configured to store one or more programs, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the video enhancement method described in any one of claims 1-7.

10. A non-volatile storage medium storing computer executable instructions, wherein the computer executable instructions are configured to perform the video enhancement method according to any one of claims 1 to 7 when executed by a computer processor.

11. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the video enhancement method according to any one of claims 1 to 7 is implemented.

Images