Big data visualization-based film and television virtual production scheme intelligent evaluation method and system

Abstract

The invention provides an intelligent evaluation method and system for a film and television virtual production scheme based on big data visualization. The method comprises the following steps: generating an initial video based on demand information; when a user side carries out circling interaction on any initial image frame forming the initial video, carrying out image interception on the initial image frame based on a circling area corresponding to the circling interaction to obtain a target sub-image; an image modification strategy is determined based on an obtained modification mode corresponding to the user side, a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy, and the modification mode comprises manual modification and automatic modification; and carrying out association modification on each modification association frame in the initial video based on the modification sub-image to obtain an updated initial video, and carrying out image comparison on the modification sub-image and the target sub-image to obtain a modification evaluation value corresponding to the modification sub-image. According to the invention, the adaptability of diversified modification requirements is at least improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a film and television virtual production scheme intelligent evaluation method and system based on big data visualization. BACKGROUND

[0002] At present, the demand for digital content creation continues to rise, and the technology for automatically generating initial videos based on user demand information has become mature. Currently, video generation technology can quickly integrate images, audio, special effects and other elements by analyzing user input text descriptions, style preferences, material requirements and other information, and output initial videos that meet basic needs. However, such generation technology often focuses on the construction process from scratch, and lacks sufficient support for personalized adjustment and optimization of generated videos, resulting in problems such as local picture defects and content deviation from user expectations in initial videos.

[0003] For the modification needs of initial videos, existing technologies focus on single modification paths and have not formed a flexible and associated processing system. For example, Chinese patent application number CN202410156789.2 "A video single frame image fast modification method" has the core logic that the user needs to find the target frame by frame by frame browsing the initial video, and after selecting the area to be modified by the fixed size of the selection box tool, only the limited automatic modification functions such as "brightness adjustment" and "contrast enhancement" can be selected, and after modification, the user needs to manually confirm and save the single frame, and then switch to the next frame to repeat the operation. Although this scheme realizes the basic single frame modification function, the modification method is fixed, only automatic modification is supported, and the range of options is narrow, which cannot meet the user's demand for manual fine adjustment; at the same time, the original design of this scheme is completely around "single frame independent modification", and it does not consider the correlation between video frames. For the same modification needs in continuous multiple frames, there is no any linkage processing mechanism, which leads to very low modification efficiency.

[0004] The core shortcoming of existing modification technology is that when the modified area appears repeatedly in the associated image frames after the selected frame, the user needs to perform the selection and modification operation frame by frame, which not only increases the workload, but also easily leads to inconsistent modification effects of each frame, affecting the final video effect. SUMMARY

[0005] Based on the above problems, the present application is proposed to provide a film and television virtual production scheme intelligent evaluation method and system based on big data visualization to overcome the above problems or at least partially solve the above problems.

[0006] According to one aspect of the present application, a film and television virtual production scheme intelligent evaluation method based on big data visualization is provided, comprising the following steps: generating an initial video based on the obtained demand information corresponding to the user end; Based on the user's selection interaction on any initial image frame that makes up the initial video, the initial image frame is cropped based on the selection area corresponding to the selection interaction to obtain the target sub-image. Based on the obtained modification method corresponding to the user terminal, an image modification strategy is determined, and a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy. The modification method includes manual modification and automatic modification. Based on the modified sub-image, each modified associated frame in the initial video is modified to obtain an updated initial video. The modified sub-image is then compared with the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image.

[0007] Optionally, in the method according to the present invention, an image modification strategy is determined based on the acquired modification method corresponding to the user terminal, and a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy, wherein the modification method includes manual modification and automatic modification, including: Retrieve a preset mode layer, wherein the preset mode layer includes each modification slot for filling each modification attribute, wherein the modification attributes include automatic modification and manual modification; Based on the user's interaction with any modification slot in the preset mode layer, the modification mode corresponding to the modification slot is determined as the modification mode corresponding to the user. If the modification method corresponding to the user terminal is manual modification, the target sub-image is sent to the user terminal, and the modified sub-image corresponding to the target sub-image uploaded by the user terminal is obtained; If the modification method corresponding to the user is automatic modification, the target sub-image is updated based on the determined region modification strategy to obtain the modified sub-image.

[0008] Optionally, in the method according to the present invention, if the modification mode corresponding to the user terminal is automatic modification, updating the target sub-image based on a determined region modification strategy to obtain a modified sub-image includes: If the modification method corresponding to the user terminal is automatic modification, perform image recognition on the target sub-image to obtain each target element located in the target sub-image; Based on each target element, the target sub-image is divided into regions to obtain each target region corresponding to each target element and the remaining background regions. Based on the user's interaction with any region located in the target sub-image, the region is determined as the target region; Retrieve a preset attribute layer, wherein the preset attribute layer includes attribute slots for filling each modified attribute, wherein the modified attributes include size change attributes, added attributes, and deleted attributes; Based on the user's interaction with any attribute slot in the preset attribute layer, the modified attribute corresponding to the attribute slot is determined as the regional attribute corresponding to the target region. The target sub-image is updated based on the region modification strategy determined by the region attributes corresponding to the target region, resulting in a modified sub-image.

[0009] Optionally, in the method according to the present invention, updating the target sub-image based on a region modification strategy determined according to the region attributes corresponding to the target region to obtain a modified sub-image includes: If the region attribute corresponding to the target region is a size change attribute or a new attribute, a preset modification layer is retrieved, wherein the preset modification layer includes each text fill slot; Retrieve the text corresponding to the filling operations performed by the user on each text fill slot based on the preset modified layer; Semantic recognition is performed on each filled text to obtain each modification information corresponding to the user terminal, wherein the modification information includes, but is not limited to, size information and element information; The target sub-image is updated based on the modified information to obtain the modified sub-image.

[0010] Optionally, in the method according to the invention, the method further includes: If the region attribute corresponding to the target region is a deletion attribute, determine whether any background region exists in the target sub-image; If any background region exists in the target sub-image, determine the positional distance between each background pixel in each background region and the target region, and sort the background regions from smallest to largest based on the positional distance to obtain a region sequence; The target area corresponding to the target region is multiplied by a preset area coefficient to obtain an area threshold, and the background area corresponding to each background region is determined sequentially based on the region sequence. If the area of ​​any corresponding background region is greater than the area threshold, the background region is determined as the control region, and the target sub-image is updated based on the control region to obtain the modified sub-image.

[0011] Optionally, in the method according to the invention, the method further includes: If there is no background region in the target sub-image and / or the area of ​​each corresponding background region is less than or equal to the area threshold, the initial image frame corresponding to the target sub-image is determined based on the initial video. Determine the region contour corresponding to the selected area in the initial image frame, and enlarge the region contour based on the retrieved preset magnification factor; The process continues until any background region exists in the updated target sub-image, and the area of ​​the region containing any corresponding background region is greater than the area threshold.

[0012] Optionally, in the method according to the invention, updating the target sub-image based on the comparison region to obtain the modified sub-image includes: Determine the value of each reference pixel in each reference pixel that makes up the reference region, and determine whether there is a pixel change trend in the horizontal and vertical directions of the reference region in the corresponding target sub-image based on each reference pixel value; If a pixel change trend exists, determine the trend direction corresponding to the pixel change trend and each target pixel in the target area, and generate a pixel trend line that passes through each target pixel and is parallel to the trend direction. Based on the trend direction, sort the target pixels corresponding to each pixel trend line to obtain each pixel sequence; Based on the distance between the points corresponding to the control area and the pixel change trend, the modified pixel values ​​of each target pixel point in each pixel sequence are determined sequentially to obtain the modified sub-image; If there is no pixel change trend, the pixel values ​​of each target pixel point corresponding to the target region are updated based on the comparison pixel values ​​to obtain the modified sub-image.

[0013] Optionally, in the method according to the invention, updating the initial video based on the modified sub-image includes: Based on the initial video, the initial image frame of the corresponding target sub-image is determined as the starting image frame, and the starting image frame and each initial image frame after the starting image frame are divided into an image determination group. Image recognition is performed on each initial image frame in the image determination group, and based on the recognition results, it is determined whether there are associated regions in each initial image frame that have the same image content as the target sub-image; Each initial image frame with associated regions is identified as a modified associated frame, and each associated region in each modified associated frame is replaced based on the modified sub-image to obtain the updated modified associated frames. The initial video is updated based on the updated modified associated frames.

[0014] Optionally, in the method according to the present invention, image comparison is performed between the modified sub-image and the target sub-image to obtain a modification evaluation value corresponding to the modified sub-image, including: The modified sub-image and the target sub-image are binarized to obtain a first image and a second image. The first image includes each first image pixel corresponding to the first pixel value, and the second image includes each second image pixel corresponding to the second pixel value. Create a transparent blending layer, and then stack the first image and the second image on the transparent blending layer in sequence; The first and second numbers of each first image pixel and each second image pixel located in the transparent blending layer are summed, and the resulting number of pixels is multiplied by a preset pixel coefficient to obtain the pixel evaluation value. The association evaluation value is obtained by multiplying the determined number of associations for each modified association frame with the preset association coefficient. The modification evaluation value is obtained by summing the pixel evaluation values ​​and associated evaluation values ​​corresponding to the same modified sub-image.

[0015] According to another aspect of the present invention, an intelligent evaluation system for film and television virtual production schemes based on big data visualization is provided, comprising: The generation module is configured to generate an initial video based on the acquired demand information corresponding to the user's needs. The interaction module is configured to perform image cropping on the initial image frame based on the selection interaction performed by the user terminal on any initial image frame that makes up the initial video, and to obtain the target sub-image based on the selection area corresponding to the selection interaction. The determination module is configured to determine an image modification strategy based on the obtained modification method corresponding to the user terminal, and generate a modified sub-image corresponding to the target sub-image based on the image modification strategy, wherein the modification method includes manual modification and automatic modification; The evaluation module is configured to perform associated modifications on each modified associated frame in the initial video based on the modified sub-image to obtain the updated initial video, and to perform image comparison between the modified sub-image and the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image.

[0016] According to the present invention, firstly, it enables precise and personalized adaptation of video modification. By accurately responding to the client's selection interaction to locate the target sub-image, and flexibly matching the corresponding image modification strategies for both manual and automatic modification methods, it not only meets the user's need for fine-tuning of details but also improves the processing efficiency of conventional modification scenarios, thereby effectively adapting to diverse video optimization demands. Secondly, it significantly improves the coherence and convenience of continuous video modification. By synchronously modifying each related frame in the initial video, it not only avoids the tedious process of users operating on recurring target areas frame by frame and reduces the cost of manual intervention but also ensures the consistency of modification effects of each related frame, helping to improve the overall quality of video content. Finally, the solution compares the modified sub-image with the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image. The modification evaluation value can intuitively reflect the actual degree of modification made by the user to the initial video, thereby accurately judging whether the generated effect of the initial video meets the user's expectations and providing data support for subsequent video generation to be closer to the user's expected effect. Attached Figure Description

[0017] Figure 1 A flowchart of an intelligent evaluation method for film and television virtual production schemes based on big data visualization, according to an embodiment of the present invention, is shown. Figure 2 A structural block diagram of an intelligent evaluation system for film and television virtual production schemes based on big data visualization, according to another embodiment of the present invention, is shown. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] To address the problems existing in the aforementioned background technology, the inventors have proposed the solution of this invention. One embodiment of this invention provides an intelligent evaluation method for film and television virtual production schemes based on big data visualization, which can be executed on a computing device.

[0020] Figure 1 A flowchart of an intelligent evaluation method for film and television virtual production schemes based on big data visualization, according to an embodiment of the present invention, is shown. This method is suitable for execution in a computing device.

[0021] like Figure 1As shown, the intelligent evaluation method for film and television virtual production schemes based on big data visualization proposed in this embodiment begins with step S102, which includes the following: An initial video is generated based on the acquired demand information corresponding to the user's needs. Based on the user's selection interaction on any initial image frame that makes up the initial video, the initial image frame is cropped based on the selection area corresponding to the selection interaction to obtain the target sub-image. Based on the obtained modification method corresponding to the user terminal, an image modification strategy is determined, and a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy. The modification method includes manual modification and automatic modification. Based on the modified sub-image, each modified associated frame in the initial video is modified to obtain an updated initial video. The modified sub-image is then compared with the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image.

[0022] Optionally, in the method according to the present invention, an image modification strategy is determined based on the acquired modification method corresponding to the user terminal, and a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy, wherein the modification method includes manual modification and automatic modification, including: Retrieve a preset mode layer, wherein the preset mode layer includes each modification slot for filling each modification attribute, wherein the modification attributes include automatic modification and manual modification; Based on the user's interaction with any modification slot in the preset mode layer, the modification mode corresponding to the modification slot is determined as the modification mode corresponding to the user. If the modification method corresponding to the user terminal is manual modification, the target sub-image is sent to the user terminal, and the modified sub-image corresponding to the target sub-image uploaded by the user terminal is obtained; If the modification method corresponding to the user is automatic modification, the target sub-image is updated based on the determined region modification strategy to obtain the modified sub-image.

[0023] Optionally, in the method according to the present invention, if the modification mode corresponding to the user terminal is automatic modification, updating the target sub-image based on a determined region modification strategy to obtain a modified sub-image includes: If the modification method corresponding to the user terminal is automatic modification, perform image recognition on the target sub-image to obtain each target element located in the target sub-image; Based on each target element, the target sub-image is divided into regions to obtain each target region corresponding to each target element and the remaining background regions. Based on the user's interaction with any region located in the target sub-image, the region is determined as the target region; Retrieve a preset attribute layer, wherein the preset attribute layer includes attribute slots for filling each modified attribute, wherein the modified attributes include size change attributes, added attributes, and deleted attributes; Based on the user's interaction with any attribute slot in the preset attribute layer, the modified attribute corresponding to the attribute slot is determined as the regional attribute corresponding to the target region. The target sub-image is updated based on the region modification strategy determined by the region attributes corresponding to the target region, resulting in a modified sub-image.

[0024] Optionally, in the method according to the present invention, updating the target sub-image based on a region modification strategy determined according to the region attributes corresponding to the target region to obtain a modified sub-image includes: If the region attribute corresponding to the target region is a size change attribute or a new attribute, a preset modification layer is retrieved, wherein the preset modification layer includes each text fill slot; Retrieve the text corresponding to the filling operations performed by the user on each text fill slot based on the preset modified layer; Semantic recognition is performed on each filled text to obtain each modification information corresponding to the user terminal, wherein the modification information includes, but is not limited to, size information and element information; The target sub-image is updated based on the modified information to obtain the modified sub-image.

[0025] Optionally, in the method according to the invention, the method further includes: If the region attribute corresponding to the target region is a deletion attribute, determine whether any background region exists in the target sub-image; If any background region exists in the target sub-image, determine the positional distance between each background pixel in each background region and the target region, and sort the background regions from smallest to largest based on the positional distance to obtain a region sequence; The target area corresponding to the target region is multiplied by a preset area coefficient to obtain an area threshold, and the background area corresponding to each background region is determined sequentially based on the region sequence. If the area of ​​any corresponding background region is greater than the area threshold, the background region is determined as the control region, and the target sub-image is updated based on the control region to obtain the modified sub-image.

[0026] Optionally, in the method according to the invention, the method further includes: If there is no background region in the target sub-image and / or the area of ​​each corresponding background region is less than or equal to the area threshold, the initial image frame corresponding to the target sub-image is determined based on the initial video. Determine the region contour corresponding to the selected area in the initial image frame, and enlarge the region contour based on the retrieved preset magnification factor; The process continues until any background region exists in the updated target sub-image, and the area of ​​the region containing any corresponding background region is greater than the area threshold.

[0027] Optionally, in the method according to the invention, updating the target sub-image based on the comparison region to obtain the modified sub-image includes: Determine the value of each reference pixel in each reference pixel that makes up the reference region, and determine whether there is a pixel change trend in the horizontal and vertical directions of the reference region in the corresponding target sub-image based on each reference pixel value; If a pixel change trend exists, determine the trend direction corresponding to the pixel change trend and each target pixel in the target area, and generate a pixel trend line that passes through each target pixel and is parallel to the trend direction. Based on the trend direction, sort the target pixels corresponding to each pixel trend line to obtain each pixel sequence; Based on the distance between the points corresponding to the control area and the pixel change trend, the modified pixel values ​​of each target pixel point in each pixel sequence are determined sequentially to obtain the modified sub-image; If there is no pixel change trend, the pixel values ​​of each target pixel point corresponding to the target region are updated based on the comparison pixel values ​​to obtain the modified sub-image.

[0028] Optionally, in the method according to the invention, updating the initial video based on the modified sub-image includes: Based on the initial video, the initial image frame of the corresponding target sub-image is determined as the starting image frame, and the starting image frame and each initial image frame after the starting image frame are divided into an image determination group. Image recognition is performed on each initial image frame in the image determination group, and based on the recognition results, it is determined whether there are associated regions in each initial image frame that have the same image content as the target sub-image; Each initial image frame with associated regions is identified as a modified associated frame, and each associated region in each modified associated frame is replaced based on the modified sub-image to obtain the updated modified associated frames. The initial video is updated based on the updated modified associated frames.

[0029] Optionally, in the method according to the present invention, image comparison is performed between the modified sub-image and the target sub-image to obtain a modification evaluation value corresponding to the modified sub-image, including: The modified sub-image and the target sub-image are binarized to obtain a first image and a second image. The first image includes each first image pixel corresponding to the first pixel value, and the second image includes each second image pixel corresponding to the second pixel value. Create a transparent blending layer, and then stack the first image and the second image on the transparent blending layer in sequence; The first and second numbers of each first image pixel and each second image pixel located in the transparent blending layer are summed, and the resulting number of pixels is multiplied by a preset pixel coefficient to obtain the pixel evaluation value. The association evaluation value is obtained by multiplying the determined number of associations for each modified association frame with the preset association coefficient. The modification evaluation value is obtained by summing the pixel evaluation values ​​and associated evaluation values ​​corresponding to the same modified sub-image.

[0030] According to the present invention, firstly, it enables precise and personalized adaptation of video modification. By accurately responding to the client's selection interaction to locate the target sub-image, and flexibly matching the corresponding image modification strategies for both manual and automatic modification methods, it not only meets the user's need for fine-tuning of details but also improves the processing efficiency of conventional modification scenarios, thereby effectively adapting to diverse video optimization demands. Secondly, it significantly improves the coherence and convenience of continuous video modification. By synchronously modifying each related frame in the initial video, it not only avoids the tedious process of users operating on recurring target areas frame by frame and reduces the cost of manual intervention but also ensures the consistency of modification effects of each related frame, helping to improve the overall quality of video content. Finally, the solution compares the modified sub-image with the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image. The modification evaluation value can intuitively reflect the actual degree of modification made by the user to the initial video, thereby accurately judging whether the generated effect of the initial video meets the user's expectations and providing data support for subsequent video generation to be closer to the user's expected effect.

[0031] Another embodiment of the present invention provides an intelligent evaluation system for film and television virtual production schemes based on big data visualization. Figure 2 Its corresponding system block diagram includes: The generation module is configured to generate an initial video based on the acquired demand information corresponding to the user's needs. The interaction module is configured to perform image cropping on the initial image frame based on the selection interaction performed by the user terminal on any initial image frame that makes up the initial video, and to obtain the target sub-image based on the selection area corresponding to the selection interaction. The determination module is configured to determine an image modification strategy based on the obtained modification method corresponding to the user terminal, and generate a modified sub-image corresponding to the target sub-image based on the image modification strategy, wherein the modification method includes manual modification and automatic modification; The evaluation module is configured to perform associated modifications on each modified associated frame in the initial video based on the modified sub-image to obtain the updated initial video, and to perform image comparison between the modified sub-image and the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image.

[0032] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing preferred embodiments of the invention.

[0033] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0034] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0035] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0036] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components.

[0037] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0038] Furthermore, some of the embodiments described herein are methods or combinations of method elements that can be implemented by a processor of a computer system or by other means of performing the functions. Therefore, a processor having the necessary instructions for implementing the methods or method elements forms means for implementing the methods or method elements. Furthermore, the elements described herein in the apparatus embodiments are examples of means for implementing the functions performed by elements for the purposes of carrying out the invention.

[0039] As used herein, unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

[0040] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and edibility purposes, and not for the purpose of explaining or limiting the subject matter of the invention.

Claims

1. A method for intelligent evaluation of film and television virtual production schemes based on big data visualization, characterized in that, include: An initial video is generated based on the acquired demand information corresponding to the user's needs. When a user interacts with the selection of any initial image frame that makes up the initial video, the initial image frame is cropped based on the selection area corresponding to the selection interaction to obtain the target sub-image. Based on the obtained modification method corresponding to the user terminal, an image modification strategy is determined, and a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy. The modification method includes manual modification and automatic modification. Based on the modified sub-image, each modified associated frame in the initial video is modified to obtain an updated initial video. The modified sub-image is then compared with the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image.

2. The method according to claim 1, characterized in that, An image modification strategy is determined based on the obtained modification method corresponding to the user terminal, and a modified sub-image corresponding to the target sub-image is generated based on the image modification strategy. The modification method includes manual modification and automatic modification, including: Retrieve a preset mode layer, wherein the preset mode layer includes each modification slot for filling each modification attribute, wherein the modification attributes include automatic modification and manual modification; Based on the user's interaction with any modification slot in the preset mode layer, the modification mode corresponding to the modification slot is determined as the modification mode corresponding to the user. If the modification method corresponding to the user terminal is manual modification, the target sub-image is sent to the user terminal, and the modified sub-image corresponding to the target sub-image uploaded by the user terminal is obtained; If the modification method corresponding to the user is automatic modification, the target sub-image is updated based on the determined region modification strategy to obtain the modified sub-image.

3. The method according to claim 2, characterized in that, If the modification method corresponding to the user terminal is automatic modification, the target sub-image is updated based on a determined region modification strategy to obtain the modified sub-image, including: If the modification method corresponding to the user terminal is automatic modification, perform image recognition on the target sub-image to obtain each target element located in the target sub-image; Based on each target element, the target sub-image is divided into regions to obtain each target region corresponding to each target element and the remaining background regions. Based on the user's interaction with any region located in the target sub-image, the region is determined as the target region; Retrieve a preset attribute layer, wherein the preset attribute layer includes attribute slots for filling each modified attribute, wherein the modified attributes include size change attributes, added attributes, and deleted attributes; Based on the user's interaction with any attribute slot in the preset attribute layer, the modified attribute corresponding to the attribute slot is determined as the regional attribute corresponding to the target region. The target sub-image is updated based on the region modification strategy determined by the region attributes corresponding to the target region, resulting in a modified sub-image.

4. The method according to claim 3, characterized in that, The target sub-image is updated based on a region modification strategy determined according to the region attributes corresponding to the target region, resulting in a modified sub-image, including: If the region attribute corresponding to the target region is a size change attribute or a new attribute, a preset modification layer is retrieved, wherein the preset modification layer includes each text fill slot; Retrieve the text corresponding to the filling operations performed by the user on each text fill slot based on the preset modified layer; Semantic recognition is performed on each filled text to obtain each modification information corresponding to the user terminal, wherein the modification information includes, but is not limited to, size information and element information; The target sub-image is updated based on the modified information to obtain the modified sub-image.

5. The method according to claim 4, characterized in that, The method further includes: If the region attribute corresponding to the target region is a deletion attribute, determine whether any background region exists in the target sub-image; If any background region exists in the target sub-image, determine the positional distance between each background pixel in each background region and the target region, and sort the background regions from smallest to largest based on the positional distance to obtain a region sequence; The target area corresponding to the target region is multiplied by a preset area coefficient to obtain an area threshold, and the background area corresponding to each background region is determined sequentially based on the region sequence. If the area of ​​any corresponding background region is greater than the area threshold, the background region is determined as the control region, and the target sub-image is updated based on the control region to obtain the modified sub-image.

6. The method according to claim 5, characterized in that, The method further includes: If there is no background region in the target sub-image and / or the area of ​​each corresponding background region is less than or equal to the area threshold, the initial image frame corresponding to the target sub-image is determined based on the initial video. Determine the region contour corresponding to the selected area in the initial image frame, and enlarge the region contour based on the retrieved preset magnification factor; The process continues until any background region exists in the updated target sub-image, and the area of ​​the region containing any corresponding background region is greater than the area threshold.

7. The method according to claim 5, characterized in that, The target sub-image is updated based on the comparison region to obtain the modified sub-image, including: Determine the value of each reference pixel in each reference pixel that makes up the reference region, and determine whether there is a pixel change trend in the horizontal and vertical directions of the reference region in the corresponding target sub-image based on each reference pixel value; If a pixel change trend exists, determine the trend direction corresponding to the pixel change trend and each target pixel in the target area, and generate a pixel trend line that passes through each target pixel and is parallel to the trend direction. Based on the trend direction, sort the target pixels corresponding to each pixel trend line to obtain each pixel sequence; Based on the distance between the points corresponding to the control area and the pixel change trend, the modified pixel values ​​of each target pixel point in each pixel sequence are determined sequentially to obtain the modified sub-image; If there is no pixel change trend, the pixel values ​​of each target pixel point corresponding to the target region are updated based on the comparison pixel values ​​to obtain the modified sub-image.

8. The method according to claim 7, characterized in that, Based on the modified sub-images, each modified associated frame in the initial video is modified accordingly to obtain the updated initial video, including: Based on the initial video, the initial image frame of the corresponding target sub-image is determined as the starting image frame, and the starting image frame and each initial image frame after the starting image frame are divided into an image determination group. Image recognition is performed on each initial image frame in the image determination group, and based on the recognition results, it is determined whether there are associated regions in each initial image frame that have the same image content as the target sub-image; Each initial image frame with associated regions is identified as a modified associated frame, and each associated region in each modified associated frame is replaced based on the modified sub-image to obtain the updated modified associated frames. The initial video is updated based on the updated modified associated frames.

9. The method according to claim 8, characterized in that, The modified sub-image is compared with the target sub-image to obtain a modification evaluation value corresponding to the modified sub-image, including: The modified sub-image and the target sub-image are binarized to obtain a first image and a second image. The first image includes each first image pixel corresponding to the first pixel value, and the second image includes each second image pixel corresponding to the second pixel value. Create a transparent blending layer, and then stack the first image and the second image on the transparent blending layer in sequence; The first and second numbers of each first image pixel and each second image pixel located in the transparent blending layer are summed, and the resulting number of pixels is multiplied by a preset pixel coefficient to obtain the pixel evaluation value. The association evaluation value is obtained by multiplying the determined number of associations for each modified association frame with the preset association coefficient. The modification evaluation value is obtained by summing the pixel evaluation values ​​and associated evaluation values ​​corresponding to the same modified sub-image.

10. An intelligent evaluation system for film and television virtual production schemes based on big data visualization, characterized in that, include: The generation module is configured to generate an initial video based on the acquired demand information corresponding to the user's needs. The interaction module is configured to perform image cropping on the initial image frame based on the selection interaction performed by the user terminal on any initial image frame that makes up the initial video, and to obtain the target sub-image based on the selection area corresponding to the selection interaction. The determination module is configured to determine an image modification strategy based on the obtained modification method corresponding to the user terminal, and generate a modified sub-image corresponding to the target sub-image based on the image modification strategy, wherein the modification method includes manual modification and automatic modification; The evaluation module is configured to perform associated modifications on each modified associated frame in the initial video based on the modified sub-image to obtain the updated initial video, and to perform image comparison between the modified sub-image and the target sub-image to obtain the modification evaluation value corresponding to the modified sub-image.

Images