Intelligent generation method and system of multimedia content based on massive material library

Abstract

The present invention discloses a method and system for intelligently generating multimedia content based on a massive material library, and relates to the technical field of multimedia content generation, and specifically includes the following steps: obtaining multiple materials to be combined from a massive material library, and collecting the lighting condition data and shooting equipment data of each material, and transmitting these data to a central processing system for storage and monitoring; analyzing the lighting condition data transmitted to the central processing system, and evaluating whether there will be a problem of inconsistent hue in the material combination process; in the case of evaluating that there is a problem of inconsistent hue in the material combination process, further analyzing the collected shooting equipment data, identifying and marking the materials with inconsistent hue; applying an optimization model to the materials marked as inconsistent hue, and recombining to generate multimedia content. The present invention solves the problem of visual disharmony caused by inconsistent hue of materials in the process of multimedia content generation, and realizes hue consistency control.

Description

Technical Field

[0001] The present invention relates to the technical field of multimedia content generation, and in particular to a method and system for intelligently generating multimedia content based on a massive material library. Background Art

[0002] Multimedia content refers to a combination of digital data that contains a variety of information presentation forms, including but not limited to text, images, audio, video, animation, interactive content, etc. These contents are usually processed and displayed through computer software or hardware systems, and can be presented to users in multiple sensory forms such as vision and hearing. Multimedia content can be generated in a variety of ways, and common generation methods include manual editing, template production, material combination, and algorithm-based automatic generation. Especially in today's information and digital context, the method of automatically generating multimedia content has been widely used. Through pre-set rules or intelligent algorithms, multimedia content with specific themes or styles can be automatically selected, combined and generated from massive material libraries, greatly improving the efficiency and personalization of content production.

[0003] Existing intelligent generation technologies for multimedia content based on massive material libraries usually rely on a variety of advanced algorithms and technologies, including machine learning, natural language processing, image recognition, video analysis, etc. First, the system classifies and labels various contents in the massive material library through indexing and classification technology for rapid retrieval and matching. Then, after the user enters the requirements or generates rules, the intelligent generation system uses recommendation algorithms to select materials related to the requirements from the material library, including text, images, audio and video clips, etc. Next, the system automatically combines these materials according to the preset content generation template or dynamically generated content structure to ensure that they form a coherent whole in terms of vision and semantics. At the same time, some systems also use deep learning models for content optimization, such as adjusting image resolution, improving audio quality, or generating natural and fluent text to achieve high-quality output effects. Finally, the multimedia content generated by the system can be further personalized according to specific application scenarios to meet the specific needs of users or optimize user experience.

[0004] The prior art has the following deficiencies:

[0005] When generating multimedia content based on a massive library of materials, existing technologies usually require combining multiple materials. However, when these materials come from different lighting conditions or use different shooting equipment, due to the inconsistency of the background tones in the library, existing technologies often find it difficult to accurately adjust the tones when automatically combining these materials, resulting in visually inconsistent content. This inconsistency in tones can make viewers feel awkward when watching content such as brand promotional videos, which in turn affects the consistency of the brand image, reduces the effectiveness of promotion, and ultimately significantly affects the professionalism and audience acceptance of the generated content.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a method and system for intelligently generating multimedia content based on a massive material library to solve the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a method for intelligently generating multimedia content based on a massive material library, specifically comprising the following steps:

[0009] Obtain multiple materials to be combined from the massive material library, collect the lighting condition data and shooting equipment data of each material, and transmit these data to the central processing system for storage and monitoring;

[0010] Analyze the lighting condition data transmitted to the central processing system to evaluate whether there will be any color inconsistencies during the material combination process;

[0011] When it is assessed that there is a problem of inconsistent color tones in the material combination process, further analyze the collected shooting equipment data, identify and mark the materials with inconsistent color tones;

[0012] Applying an optimization model to adjust materials marked as inconsistent in tone, and recombining to generate multimedia content;

[0013] The color consistency is verified based on the generated content. If a color inconsistency is detected, it is readjusted until the consistency reaches the preset standard.

[0014] Preferably, analyzing the illumination condition data transmitted to the central processing system to evaluate whether there will be a problem of inconsistent tones during the material combination process specifically includes the following steps:

[0015] Extract the lighting condition data of each material transmitted to the central processing system, including three types of data: average brightness, exposure value and color temperature;

[0016] Preprocess the three types of data, namely the average brightness, exposure value, and color temperature, for each extracted material;

[0017] Analyze the three types of data, namely the average brightness, exposure value, and color temperature, for each preprocessed material, and generate a light consistency coefficient and a color temperature balance index respectively;

[0018] Construct a hue consistency evaluation model using the generated light consistency coefficient and color temperature balance index, generate a hue consistency coefficient, and compare the generated hue consistency coefficient with a preset hue consistency threshold to evaluate whether there will be a problem of inconsistent hues during the material combination process.

[0019] Preferably, the acquisition logic of the light consistency coefficient and the color temperature balance index is as follows:

[0020] Extract the average brightness, exposure value, and color temperature of each preprocessed material, and calibrate the average brightness, exposure value, and color temperature of each preprocessed material as L m , E m , and T m , where L m represents the average brightness of the m-th preprocessed material, E m represents the exposure value of the m-th preprocessed material, T m represents the color temperature of the m-th preprocessed material, and m = 1, 2, 3,..., k, where k is a positive integer;

[0021] Compare the average brightness L i of the i-th material with the average brightness L j of the j-th material, where the value ranges of i and j are 1 ≤ i < j ≤ k, and both i and j are positive integers. Calculate the brightness difference value between the i-th material and the j-th material. The specific calculation formula is as follows:

[0022] ΔL ij = |L i - L j |

[0023] In the formula, ΔL ij is the brightness difference value between the i-th material and the j-th material;

[0024] Compare the exposure value E i of the i-th material with the exposure value E j of the j-th material, and calculate the exposure difference value between the i-th material and the j-th material. The specific calculation formula is as follows:

[0025] ΔE ij = |E i - E j |

[0026] In the formula, ΔE ij is the exposure difference between the i-th material and the j-th material;

[0027] Calculate the exposure adjustment factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0028]

[0029] In the formula, is the exposure adjustment factor between the i-th material and the j-th material;

[0030] The brightness difference between the i-th material and the j-th material is corrected by combining the exposure adjustment factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0031]

[0032] In the formula, is the brightness difference between the i-th material and the j-th material after correction;

[0033] Calculate the lighting consistency coefficient. The specific calculation formula is as follows:

[0034]

[0035] In the formula, C L is the illumination consistency coefficient;

[0036] The color temperature T of the i-th material i and the color temperature T of the jth material j Compare and calculate the color temperature difference between the i-th material and the j-th material. The specific calculation formula is as follows:

[0037] ΔT ij =|T i -T j |

[0038] In the formula, ΔT ij is the color temperature difference between the i-th material and the j-th material;

[0039] Calculate the smoothing factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0040]

[0041] In the formula, is the smoothing factor between the i-th material and the j-th material;

[0042] The color temperature difference between the i-th material and the j-th material is corrected by combining the smoothing factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0043]

[0044] In the formula, is the color temperature difference between the i-th material and the j-th material after correction;

[0045] Calculate the color temperature balance index. The specific calculation formula is as follows:

[0046]

[0047] In the formula, I T It is the color temperature balance index.

[0048] Preferably, the generated illumination consistency coefficient C L and color temperature balance index I T Construct a hue consistency evaluation model, generate a hue consistency coefficient CTCC by weighted summation, and compare the generated hue consistency coefficient CTCC with the preset hue consistency threshold CTCC yz Compare and evaluate whether there will be color inconsistency problems during the material combination process. The specific analysis is as follows:

[0049] If CTCC <CTCC yz , there will be no color inconsistency during the material combination process;

[0050] If CTCC ≥ CTCC yz , there will be color inconsistency problems during the material combination process.

[0051] Preferably, when it is assessed that there is a problem of inconsistent color tone in the process of combining materials, further analyzing the collected shooting device data, identifying and marking the materials with inconsistent color tone, specifically includes the following steps:

[0052] Acquire the shooting device data associated with each material and perform preprocessing after acquisition;

[0053] Extracting shutter speed and sensitivity of each material from pre-processed camera data;

[0054] The shutter speed and sensitivity of each extracted material are standardized, and after the standardization, a comprehensive analysis is performed to generate the equipment impact coefficient of each material by building an equipment impact assessment model;

[0055] The device impact coefficient of each generated material is compared with a preset device impact threshold, and materials whose device impact coefficient exceeds the preset device impact threshold are identified and marked as materials with inconsistent tones.

[0056] Preferably, the shutter speed and sensitivity of each extracted material are standardized, and after the standardization, a comprehensive analysis is performed to generate the equipment impact coefficient of each material by constructing an equipment impact assessment model, which specifically includes the following steps:

[0057] The shutter speed and sensitivity of each extracted material are calibrated as V m and ISO m , V m Indicates the shutter speed of the mth material, ISO m represents the sensitivity of the m-th material, m=1, 2, 3, ..., k, k is a positive integer;

[0058] Get the maximum and minimum shutter speeds of all materials, and mark the maximum and minimum shutter speeds of all materials as V max and V min , get the maximum and minimum values ​​of the sensitivity of all materials, and calibrate the maximum and minimum values ​​of the sensitivity of all materials as ISO max and ISO min ;

[0059] The shutter speed V for each clip m and ISO m Perform standardization, the standardization formula is as follows:

[0060]

[0061] In the formula, is the normalized shutter speed of the mth material, is the normalized sensitivity of the mth material;

[0062] Collect several shutter speeds, several sensitivities and corresponding equipment influence coefficients of the m-th material after standardization in the past period of time, and calibrate them as and DIF m x , x represents the number of shutter speeds, sensitivities and corresponding equipment influence coefficients of the m-th material after standardization in the past period of time, x=3, 4, 5, ..., h, h is a positive integer, and the data collected in the past period of time is formed into a historical data set;

[0063] Select the multivariate regression model as the equipment impact assessment model, and train it through the historical data set to determine the value of the regression coefficient according to the formula:

[0064]

[0065] Where β0, β1 and β2 are regression coefficients;

[0066] By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined;

[0067] Using the final regression coefficient, the constructed device impact assessment model is input with the shutter speed of the mth material generated in real time. and sensitivity Generate the device influence coefficient DIF of the mth material in real time m .

[0068] Preferably, the device influence coefficient DIF of each material generated m With the pre-set device impact threshold DIF yz Compare and identify the equipment impact coefficient DIF m Exceeds the preset device impact threshold DIF yz The material is marked as inconsistent in tone. The specific analysis is as follows:

[0069] If DIF m <DIF yz , do not mark the material as tonally inconsistent material;

[0070] If DIF m ≥DIF yz , marking the clip as tonally inconsistent.

[0071] Preferably, the materials marked as inconsistent in hue are adjusted by applying an optimization model and recombined to generate multimedia content, which specifically includes the following steps: extracting the equipment influence coefficient and shooting equipment data of the materials marked as inconsistent in hue, inputting these data into the optimization model, and calculating the hue adjustment parameters, including the exposure adjustment factor and the color temperature correction coefficient; applying the hue adjustment parameters, performing image processing on the material, adjusting its brightness and color temperature to make it consistent with the lighting conditions and color temperature of other materials; finally, recombining the adjusted material with other unmarked materials to generate multimedia content.

[0072] Preferably, a multimedia content intelligent generation system based on a massive material library includes a material collection and data integration module, a lighting analysis and hue evaluation module, a hue consistency recognition and marking module, an optimization model application and material reorganization module, and a hue consistency verification and dynamic adjustment module;

[0073] The material acquisition and data integration module obtains multiple materials to be combined from the massive material library, collects the lighting condition data and shooting equipment data of each material, and transmits these data to the central processing system for storage and monitoring;

[0074] The lighting analysis and color tone evaluation module analyzes the lighting condition data transmitted to the central processing system to evaluate whether color tone inconsistencies will occur during the material combination process;

[0075] The color consistency recognition and marking module further analyzes the collected shooting equipment data, identifies the materials with inconsistent colors and marks them when it is assessed that there are color inconsistencies in the material combination process;

[0076] The optimization model application and material reorganization module adjusts the materials marked as inconsistent in tone by applying the optimization model and recombines them to generate multimedia content;

[0077] The tone consistency verification and dynamic adjustment module verifies the tone consistency based on the generated content. If tone inconsistency is detected, it is readjusted until the consistency reaches the preset standard.

[0078] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0079] 1. The present invention improves the consistency of tones during the multimedia content generation process, thereby ensuring that the final generated content appears visually professional and coordinated. By comprehensively acquiring and analyzing the lighting condition data and shooting equipment data during the material collection stage, the system can accurately assess the tones inconsistency that may occur during the material combination process. This early prediction mechanism greatly reduces the need for large-scale corrections to the materials after combination, making the entire processing flow more efficient and improving the quality and consistency of content generation.

[0080] 2. The present invention describes a dynamic adjustment process for problematic materials. This adjustment method based on an optimization model can identify and correct materials with inconsistent tones in real time during the material combination process, thereby largely avoiding visual abruptness in the final generated content. The system ensures that each material can be consistent with other materials after tones are adjusted through standardization, device influence coefficient calculation, and comparison with preset thresholds, so that the final output multimedia content is not only harmonious in tones, but also maintains a high level of visual expression.

[0081] 3. The present invention verifies the consistency of the color tone of the generated content, and the system realizes a closed-loop optimization process. This process ensures that even if there is still a problem of inconsistent color tone after the first adjustment, the preset color tone consistency standard can be finally achieved through repeated adjustment and verification. The advantage of this technical solution is that it provides a highly automated and accurate color tone consistency control mechanism, which not only improves the production efficiency of multimedia content, but also improves the overall visual effect and brand image consistency of high-end content such as brand promotional videos, greatly enhancing the audience's viewing experience and acceptance. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0083] Figure 1 The figure is a flow chart of the method and system for intelligently generating multimedia content based on a massive material library of the present invention.

[0084] Figure 2 It is a module schematic diagram of the method and system for intelligently generating multimedia content based on a massive material library of the present invention. DETAILED DESCRIPTION

[0085] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0086] The present invention provides Figure 1 The method for intelligently generating multimedia content based on a massive material library shown in the figure specifically includes the following steps:

[0087] Obtain multiple materials to be combined from the massive material library, collect the lighting condition data and shooting equipment data of each material, and transmit these data to the central processing system for storage and monitoring;

[0088] There are several ways to obtain multiple materials to be combined from a massive material library: First, you can use keyword retrieval or tag matching to filter out materials that meet the theme using preset tags in the material library or keywords entered by the user; second, you can use content analysis algorithms, such as image recognition, audio analysis, or video frame analysis, to automatically identify the content type and attributes of the material so that relevant materials can be extracted from it. In addition, you can also use recommendation algorithms based on user preferences and historical usage data to automatically recommend materials that are similar to previously used materials or user preferences. When obtaining materials, the system will call the API interface of the material library, extract a list of qualified materials from the material library based on the set retrieval conditions and analysis results, and mark these materials as materials to be combined for subsequent processing.

[0089] The collection of lighting condition data and shooting equipment data for each material can be achieved in the following ways: the system can use image analysis algorithms to detect the lighting condition data of each material, such as obtaining the material's average brightness, exposure value, color temperature and other lighting-related data through histogram analysis, brightness value distribution, contrast evaluation and other technologies. At the same time, by parsing the metadata of the material (such as EXIF ​​data), the system can extract the data of the shooting equipment, including but not limited to the model of the shooting equipment, sensor type, lens focal length, aperture value, ISO sensitivity, shutter speed and other parameters. These data can be automatically extracted through standard data interfaces or metadata parsing tools. After completing data collection, the system will structure the lighting condition data and shooting equipment data, store them in the database of the central processing system, and track the application of these data in the material combination process through the real-time monitoring module to ensure the accuracy and effectiveness of subsequent processing.

[0090] Analyze the lighting condition data transmitted to the central processing system to evaluate whether there will be any color inconsistencies during the material combination process;

[0091] In this embodiment, the illumination condition data transmitted to the central processing system is analyzed to evaluate whether the problem of inconsistent tones will occur during the material combination process, which specifically includes the following steps:

[0092] Extract the lighting condition data of each material transmitted to the central processing system, including three types of data: average brightness, exposure value and color temperature;

[0093] Extracting the lighting condition data (including average brightness, exposure value and color temperature) of each material transmitted to the central processing system can be achieved in the following ways: First, the system can use image processing algorithms to analyze each frame of the material. The average brightness can be extracted by calculating the brightness histogram of the image to obtain the average value of all pixel brightness values, which can be achieved by converting to a grayscale image based on the RGB model or the Y component in the YUV color space. The exposure value can be extracted by analyzing the exposure histogram of the image to identify the overall exposure level of the image, which is usually determined by combining the brightness peak position of the image and the exposure compensation data. The color temperature can be extracted by analyzing the color distribution of the image, using a white balance algorithm or a color temperature detection algorithm to infer the corresponding color temperature value from the RGB channel, which is usually calculated by analyzing the ratio of the red to blue channels. After these data are extracted, they will be uniformly formatted and stored in the database of the central processing system for subsequent analysis and processing. The entire process is completely completed by software automation without manual intervention, ensuring the efficiency and accuracy of data extraction.

[0094] Pre-process the three types of data, namely, average brightness, exposure value and color temperature of each extracted material;

[0095] The three types of data extracted by preprocessing, namely, average brightness, exposure value and color temperature, are to ensure the accuracy and consistency of these data in subsequent analysis, thereby improving the reliability of color consistency evaluation. The first step of preprocessing is to remove outliers. This step is to exclude abnormal data caused by extreme lighting conditions or erroneous data collection. This can be done by setting a reasonable threshold range to identify and eliminate data beyond the normal range. Next, the data is standardized to convert data of different ranges and units into a unified scale so that they can be directly compared and calculated in subsequent analysis. For example, the values ​​of average brightness, exposure value and color temperature are converted to a standard range of 0 to 1 through a normalization algorithm to ensure that these data have the same influence weight when calculating the lighting consistency coefficient and color temperature balance index. Finally, data smoothing can be performed to reduce the impact of noise on the data. This step is usually achieved through a moving average or a low-pass filter to ensure that the data can better reflect the actual lighting conditions and color temperature of the material when analyzing. Through these preprocessing steps, the accuracy and consistency of the data are guaranteed, laying a solid foundation for subsequent analysis and evaluation.

[0096] Analyze the average brightness, exposure value and color temperature of each pre-processed material to generate the lighting consistency coefficient and color temperature balance index respectively;

[0097] The generated illumination consistency coefficient and color temperature balance index are used to construct a hue consistency evaluation model to generate a hue consistency coefficient. The generated hue consistency coefficient is compared with a preset hue consistency threshold to evaluate whether hue inconsistency will occur during the material combination process.

[0098] The preset tone consistency threshold can be determined by analyzing a large amount of sample data, which specifically includes the following steps: First, collect and analyze the illumination consistency coefficient and color temperature balance index data of multiple sets of multimedia content that have been successfully combined. These data can be obtained from past projects or verified material libraries. Next, use statistical analysis software to perform cluster analysis and regression analysis on these data to identify the typical ranges of illumination consistency coefficient and color temperature balance index corresponding to material combinations that are visually considered to have consistent tones. Then, based on these typical ranges, set an upper limit value that can cover most successful combinations as the tone consistency threshold. Finally, through simulation tests and feedback verification of new material combinations, further fine-tune and optimize the threshold to ensure its wide applicability in different scenarios, thereby providing an accurate standard for evaluating the tone consistency of future material combinations.

[0099] In this embodiment, the logic for obtaining the illumination consistency coefficient and the color temperature balance index is as follows:

[0100] The average brightness, exposure value and color temperature of each preprocessed material are extracted, and the average brightness, exposure value and color temperature of each preprocessed material are calibrated as L m 、E m and T m , L m represents the average brightness of the mth material after preprocessing, E m represents the exposure value of the preprocessed m-th material, T m represents the color temperature of the mth material after preprocessing, m=1, 2, 3, ..., k, where k is a positive integer;

[0101] Extracting the average brightness, exposure value and color temperature of each pre-processed material can be achieved in the following ways: First, use the image processing algorithm to analyze each frame of the material. The average brightness can be extracted by calculating the brightness histogram of the image, adding up the brightness values ​​of all pixels and dividing by the total number of pixels to obtain the overall average brightness value; the exposure value E is extracted based on the metadata of the material (such as EXIF ​​information), by reading the shutter speed V, aperture value N and sensitivity ISO of the shooting device, and using the exposure value formula Calculated; the extraction of color temperature can be achieved through color space conversion. Convert the RGB image into a color space related to color temperature (such as CIE XYZ or LAB), and then use the white balance algorithm or color temperature detection algorithm to analyze the color distribution of the image and estimate the color temperature value. After these data are extracted by the algorithm, they will be structured and stored in the system for subsequent calculation of lighting consistency and color temperature balance.

[0102] Compare the average brightness L of the i-th material i with the average brightness L of the j-th material j where the ranges of i and j are 1 ≤ i < j ≤ k, and both i and j are positive integers. Calculate the brightness difference value between the i-th material and the j-th material. The specific calculation formula is as follows:

[0103] ΔL ij = |L i - L j |

[0104] In the formula, ΔL ij is the brightness difference value between the i-th material and the j-th material;

[0105] Compare the exposure value E of the i-th material i with the exposure value E of the j-th material j and calculate the exposure difference value between the i-th material and the j-th material. The specific calculation formula is as follows:

[0106] ΔE ij = |E i - E j |

[0107] In the formula, ΔE ij is the exposure difference value between the i-th material and the j-th material;

[0108] Calculate the exposure adjustment factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0109]

[0110] In the formula, is the exposure adjustment factor between the i-th material and the j-th material;

[0111] The exposure adjustment factor between the i-th material and the j-th material represents the relative difference in exposure settings between the two materials, and is used to correct the brightness difference caused by different exposure conditions. Its physical meaning is that it measures the difference between the two materials in exposure parameters such as aperture, shutter speed and ISO sensitivity. By calculating this factor, the deviation introduced by different shooting parameters can be eliminated when comparing the brightness of the materials. The purpose of this is to ensure that the comparison of brightness differences is based on real lighting conditions, rather than false differences caused by different exposure settings, thereby improving the accuracy of the calculation of the lighting consistency coefficient.

[0112] The brightness difference between the i-th material and the j-th material is corrected by combining the exposure adjustment factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0113]

[0114] In the formula, is the brightness difference between the i-th material and the j-th material after correction;

[0115] The brightness difference value is corrected in combination with the exposure adjustment factor between the i-th material and the j-th material in order to accurately reflect the actual difference in lighting conditions between the materials, rather than the error caused by different exposure settings. The shooting conditions of different materials, such as shutter speed, aperture size, and ISO sensitivity, will affect their final brightness values. If they are not adjusted, directly comparing these brightness values ​​may produce deviations due to different exposure settings. By introducing the exposure adjustment factor, the impact caused by different exposure conditions can be eliminated, so that the corrected brightness difference value more accurately represents the actual lighting difference, ensuring that the brightness comparison between the materials is fair and consistent when calculating the lighting consistency coefficient.

[0116] Calculate the lighting consistency coefficient. The specific calculation formula is as follows:

[0117]

[0118] In the formula, C L is the illumination consistency coefficient;

[0119] The size of the lighting consistency coefficient directly reflects the consistency of lighting conditions between multiple materials. The smaller the coefficient, the smaller the brightness and exposure differences between the materials, and the more consistent the lighting conditions, which means that the possibility of color inconsistency problems in the material combination process is low. On the contrary, if the lighting consistency coefficient is large, it means that there are significant brightness and exposure differences between the materials. This difference is likely to lead to inconsistent tones when combined, which in turn affects the overall visual effect of the multimedia content. Therefore, by evaluating the size of the lighting consistency coefficient, it is possible to predict the uniformity of the tones after the materials are combined, ensuring that the final generated content has good visual consistency.

[0120] The color temperature T of the i-th material i and the color temperature T of the jth material j Compare and calculate the color temperature difference between the i-th material and the j-th material. The specific calculation formula is as follows:

[0121] ΔT ij =|T i -T j

[0122] In the formula, ΔT ij is the color temperature difference between the i-th material and the j-th material;

[0123] Calculate the smoothing factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0124]

[0125] In the formula, is the smoothing factor between the i-th material and the j-th material;

[0126] The smoothing factor between the i-th material and the j-th material represents the sensitivity of the color temperature difference between the two materials, and is used to adjust the transition reaction caused by slight color temperature differences. Its physical meaning is that it measures the relative change in the color temperature of the two materials, helping to reduce the impact of small, insignificant color temperature differences. The purpose of calculating this factor is to avoid over-amplifying small color temperature differences when analyzing color temperature differences, so that the final color temperature balance index more accurately reflects the actual color consistency when the materials are combined, ensuring that the final generated multimedia content is more visually harmonious and unified.

[0127] The color temperature difference between the i-th material and the j-th material is corrected by combining the smoothing factor between the i-th material and the j-th material. The specific calculation formula is as follows:

[0128]

[0129] In the formula, is the color temperature difference between the i-th material and the j-th material after correction;

[0130] The color temperature difference value is corrected in combination with the smoothing factor between the i-th material and the j-th material in order to reduce the over-sensitivity caused by slight color temperature fluctuations or slightly different shooting conditions, so as to more accurately reflect the actual color temperature difference between the materials. In actual applications, the color temperature between two materials may have slight differences due to factors such as shooting equipment and ambient light. If they are not smoothed, these slight differences may be magnified when calculating the color temperature balance index, resulting in deviations in the final evaluation results. By introducing a smoothing factor, these slight fluctuations can be effectively filtered out, making the corrected color temperature difference value more robust, ensuring a more consistent color tone when combining materials and a more coordinated visual effect.

[0131] Calculate the color temperature balance index. The specific calculation formula is as follows:

[0132]

[0133] In the formula, I T It is the color temperature balance index.

[0134] The size of the color temperature balance index directly reflects the degree of color temperature consistency between multiple materials. The smaller the index, the smaller the color temperature difference between the materials and the more consistent the color tone, which means that the risk of color tone inconsistency during material combination is low. On the contrary, a larger color temperature balance index indicates that there is a significant color temperature difference between the materials. This difference is likely to cause inharmonious color tones when combined, which may result in abrupt visual effects. Therefore, by evaluating the size of the color temperature balance index, the consistency of the color tone after the material combination can be judged to ensure that the generated multimedia content has good visual tone uniformity and overall aesthetics.

[0135] In this embodiment, the generated illumination consistency coefficient C L and color temperature balance index I T Construct a hue consistency evaluation model, generate a hue consistency coefficient CTCC by weighted summation, and compare the generated hue consistency coefficient CTCC with the preset hue consistency threshold CTCC yz Compare and evaluate whether there will be color inconsistency problems during the material combination process. The specific analysis is as follows:

[0136] If CTCC <CTCC yz, there will be no color inconsistency during the material combination process, which means that the lighting consistency and color temperature balance of the material combination have reached the preset standards, indicating that the lighting and color tone differences between the materials are within an acceptable range. The combined multimedia content will maintain the unity of color tone visually. The resulting effect is that the generated content will not have abrupt color differences, and the audience will feel the overall coordination when watching, which helps to convey a unified visual message and enhance the coherence of the brand image;

[0137] If CTCC ≥ CTCC yz In the process of combining materials, inconsistent tones will appear, which indicates that the lighting consistency and color temperature balance of the material combination fail to meet the preset standards, suggesting that there are significant lighting or tonal differences between the materials, which may cause visual tonal inconsistencies in the combined multimedia content. The impact is that the generated content may appear unprofessional, and the audience may feel that the tones are abrupt or uncoordinated when watching, which will weaken the visual appeal of the content and the consistency of brand delivery, thereby affecting the overall visual experience and publicity effect.

[0138] When building a color consistency evaluation model, the generated lighting consistency coefficient C L and color temperature balance index I T As a key indicator, the hue consistency coefficient CTCC is generated by weighted summation. Specifically, first, C L and I T Assign weight coefficients α and β. These weight coefficients are usually determined according to the importance of each indicator in the color consistency evaluation. For example, if the consistency of lighting has a greater impact on the overall visual effect, the value of α will be relatively large. Then, through the formula CTCC = α*C L +β*I T The weighted sum is performed to generate a color consistency coefficient that comprehensively reflects the consistency of the material's illumination and color temperature. The smaller this coefficient is, the better the color consistency of the combined materials; the larger it is, the more likely there is a problem with color inconsistency.

[0139] When it is assessed that there is a problem of inconsistent color tones in the material combination process, further analyze the collected shooting equipment data, identify and mark the materials with inconsistent color tones;

[0140] In this embodiment, when it is assessed that there is a problem of inconsistent color tones in the material combination process, further analyzing the collected shooting device data, identifying and marking the materials with inconsistent color tones, specifically includes the following steps:

[0141] Acquire the shooting device data associated with each material and perform preprocessing after acquisition;

[0142] The shooting device data associated with each material can be obtained by parsing the metadata of the material file (such as EXIF ​​data). EXIF ​​data usually contains information such as shutter speed, aperture value, ISO sensitivity, shooting device model, etc. These metadata can be automatically extracted through image processing software or programming interfaces. The system first locates the metadata part of the material file, then reads and parses the data, and imports the relevant shooting parameters into the database of the central processing system. This process is fully automated and does not require human intervention, ensuring the efficiency and accuracy of data acquisition.

[0143] The purpose of preprocessing is to ensure that the acquired data is consistent and comparable in subsequent analysis. First, preprocessing can include outlier detection and elimination, eliminating extreme data caused by equipment failure or acquisition errors; second, data formatting, converting device data from different sources into a unified format for subsequent processing and analysis; finally, data standardization, converting different device parameters (such as shutter speed, ISO sensitivity, etc.) to the same dimension, eliminating the magnitude differences between different device parameters, so that data from different devices can be compared on the same scale. These preprocessing steps are implemented through automated data cleaning and format conversion tools to ensure that the processed data is accurate and unified, and can support subsequent equipment impact analysis and model building.

[0144] Extracting shutter speed and sensitivity of each material from pre-processed camera data;

[0145] Extracting the shutter speed and ISO for each clip from the preprocessed camera data can be accomplished by parsing the preprocessed data structure stored in the database. After preprocessing, the camera data is stored as a structured data record containing the relevant parameters for each clip. The system can locate and extract the corresponding shutter speed and ISO based on the unique identifier of the clip (such as the file name or ID) by querying the database. This process is automated using a database query language (such as SQL), where the system accurately extracts the shutter speed and ISO fields from the data table and uses them as input for subsequent analysis. This approach ensures accurate and efficient data extraction and can be seamlessly integrated into the entire data processing process.

[0146] The shutter speed and sensitivity of each extracted material are standardized, and after the standardization, a comprehensive analysis is performed to generate the equipment impact coefficient of each material by building an equipment impact assessment model;

[0147] The device impact coefficient of each generated material is compared with a preset device impact threshold, and materials whose device impact coefficient exceeds the preset device impact threshold are identified and marked as materials with inconsistent tones.

[0148] The pre-set device impact threshold can be determined by analyzing a large amount of historical data, which includes the following steps: First, collect and analyze multiple materials with known combination results, which have been successfully adjusted for color consistency and have good visual effects. Then, use statistical analysis tools to cluster the device impact coefficients of these materials to identify the typical range and distribution pattern of the device impact coefficient. By analyzing this data, a critical value that can cover most successful cases can be determined as the device impact threshold. In addition, the threshold can be further optimized through simulation tests and machine learning models to adapt it to the diversity of different scenarios and devices. Finally, the system uses this threshold as a benchmark for color consistency evaluation and problem identification of subsequent materials.

[0149] In this embodiment, the shutter speed and sensitivity of each extracted material are standardized, and a comprehensive analysis is performed after the standardization, and the device impact coefficient of each material is generated by constructing a device impact assessment model, which specifically includes the following steps:

[0150] The shutter speed and sensitivity of each extracted material are calibrated as V m and ISO m , V m Indicates the shutter speed of the mth material, ISO m represents the sensitivity of the m-th material, m=1, 2, 3, ..., k, k is a positive integer;

[0151] Get the maximum and minimum shutter speeds of all materials, and mark the maximum and minimum shutter speeds of all materials as V max and V min , get the maximum and minimum values ​​of the sensitivity of all materials, and calibrate the maximum and minimum values ​​of the sensitivity of all materials as ISO max and ISO min ;

[0152] The shutter speed V for each clip m and ISO m Perform standardization, the standardization formula is as follows:

[0153]

[0154] In the formula, is the normalized shutter speed of the mth material, is the normalized sensitivity of the mth material;

[0155] The purpose of standardizing the shutter speed and sensitivity of each material is to eliminate the magnitude differences in these parameters between different materials so that they can be compared on the same scale. Shutter speed and sensitivity may have large numerical differences due to different shooting equipment, shooting environment or settings, and these differences will directly affect the calculation of the material equipment impact coefficient. If standardization is not performed, the parameters of some materials may occupy too much weight in the analysis due to large or small values, resulting in deviation in the calculation results. Through standardization, the shutter speed and sensitivity data of all materials can be converted to a unified range, so that these parameters can fairly and accurately reflect their contribution to the equipment impact coefficient, ensuring the reliability and consistency of subsequent analysis.

[0156] Collect several shutter speeds, several sensitivities and corresponding equipment influence coefficients of the m-th material after standardization in the past period of time, and calibrate them as and DIF m x , x represents the number of shutter speeds, sensitivities and corresponding equipment influence coefficients of the m-th material after standardization in the past period of time, x=3, 4, 5, ..., h, h is a positive integer, and the data collected in the past period of time is formed into a historical data set;

[0157] The data recording and log management system can be used to collect several shutter speeds, sensitivities, and corresponding equipment influence coefficients of the mth material that has been standardized in the past period of time. Specifically, each time the system performs standardization and equipment impact assessment on the material, it automatically records the processing results (including standardized shutter speed, sensitivity data, and calculated equipment influence coefficient) in a central database or log file. These records usually come with timestamps and material IDs for subsequent retrieval. When it is necessary to collect these historical data, the system can query the database or log management system, extract relevant records by material ID and time range, summarize and export these data as a data set for subsequent model training and analysis. This process is fully automated, ensuring the accuracy and efficiency of data collection.

[0158] The value of x must be greater than or equal to three because in the regression model, we need to determine the specific values ​​of the three regression coefficients β0, β1, and β2. Since there is only one target equation, at least three independent equations are required to form a system of equations so that the value of each regression coefficient can be determined by solving this system of equations. Specifically, by collecting at least three independent sets of standardized shutter speed, sensitivity data and corresponding equipment influence coefficients, three independent equations can be generated so that the values ​​of the three regression coefficients can be uniquely determined. If the data is insufficient, the solution to the regression model cannot be uniquely determined, so x must be greater than or equal to three to ensure the stability and accuracy of the model.

[0159] Select the multivariate regression model as the equipment impact assessment model, and train it through the historical data set to determine the value of the regression coefficient according to the formula:

[0160]

[0161] Where β0, β1 and β2 are regression coefficients;

[0162] The reason for choosing the multiple regression model as the equipment impact assessment model is that the multiple regression model can simultaneously analyze the combined impact of multiple independent variables (such as standardized shutter speed and sensitivity) on a dependent variable (i.e., the equipment impact coefficient). This model assumes that the impact of each independent variable on the dependent variable is linearly additive, so it is suitable for evaluating the joint effects of multiple factors in complex systems. The multiple regression model can not only provide a quantitative analysis of each factor's impact coefficient on the equipment, but also quantify the relative importance of each factor by calculating the regression coefficient, thereby providing clear guidance for subsequent optimization and adjustment. Therefore, the multiple regression model is an effective and accurate method in this case, which can accurately capture the overall impact of the shooting equipment parameters on the tonal consistency.

[0163] By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized, and finally the values ​​of the regression coefficients β0, β1 and β2 are determined;

[0164] The purpose of optimizing the regression coefficients by minimizing the error between the predicted values ​​and the actual values ​​is to ensure the accuracy and predictive power of the multivariate regression model. The goal of the optimization is to find a set of regression coefficients β0, β1, and β2 so that the predicted values ​​of the model are closest to the actually observed equipment influence coefficients. The specific approach is to use the least squares method to calculate the squared error between the predicted values ​​and the actual values, and iteratively adjust the regression coefficients until the coefficient combination with the smallest error is found. This process is completed by solving the mathematical problem of minimizing the sum of squared errors, ensuring that the model has the best fit on the known data, so that it can provide more accurate predictions on the unknown data. The regression coefficients finally determined are the set of coefficients that minimize the model error, ensuring the reliability and effectiveness of the model.

[0165] Using the final regression coefficient, the constructed device impact assessment model is input with the shutter speed of the mth material generated in real time. and sensitivity Generate the device influence coefficient DIF of the mth material in real time m .

[0166] In this embodiment, the device influence coefficient DIF of each material generated m With the pre-set device impact threshold DIF yz Compare and identify the equipment impact coefficient DIF m Exceeds the preset device impact threshold DIF yz The material is marked as inconsistent in tone. The specific analysis is as follows:

[0167] If DIF m <DIF yz , not marking the material as a material with inconsistent hue means that the device influence coefficient of the material is lower than the preset threshold, indicating that the hue influence of the shooting device on the material is within an acceptable range and will not cause the phenomenon of inconsistent hue. This has a positive impact on the final generated multimedia content because the hue consistency when the materials are combined can be ensured. The marking can be achieved by adding a "consistent hue" tag to the material in the central processing system. The specific operation is that when this situation is detected, the system will automatically update the status field of the material to "consistent hue" and store it in a special material collection so that it can be used preferentially in subsequent processing;

[0168] If DIF m ≥DIF yz , marking the material as material with inconsistent hue means that the device influence coefficient of the material exceeds the preset threshold, indicating that the shooting equipment may have an adverse effect on the hue of the material, resulting in inconsistent hue, which will have a negative impact on the final generated multimedia content and requires special attention or remedial measures in subsequent processing. The marking can be achieved by adding an "inconsistent hue" label to the material in the central processing system. The specific operation is that when this situation is detected, the system will automatically update the status field of the material to "inconsistent hue" and move it to a special material collection for subsequent adjustment or further processing.

[0169] Applying an optimization model to adjust materials marked as inconsistent in tone, and recombining to generate multimedia content;

[0170] In this embodiment, the optimization model is applied to adjust the material marked as inconsistent in hue, and the material is recombined to generate multimedia content, which specifically includes the following steps: extracting the device influence coefficient and shooting device data of the material marked as inconsistent in hue, inputting these data into the optimization model, and calculating the hue adjustment parameters, including the exposure adjustment factor and the color temperature correction coefficient; applying the hue adjustment parameters, performing image processing on the material, adjusting its brightness and color temperature to make it consistent with the lighting conditions and color temperature of other materials; finally, recombining the adjusted material with other unmarked materials to generate multimedia content.

[0171] When generating multimedia content, if some materials are marked as inconsistent in color tone, it means that these materials have significant differences in brightness and color temperature from other materials. In order to eliminate these differences, the system first extracts the equipment influence coefficients and related shooting equipment data of these materials and inputs these data into an optimization model. By analyzing these data, the model calculates the color tone adjustment parameters for adjusting the brightness and color temperature of the materials, including the exposure adjustment factor and the color temperature correction coefficient. After applying these adjustment parameters, the brightness and color temperature of the materials will be corrected to make them consistent with other materials, thereby avoiding the phenomenon of inconsistent color tones in the final combined multimedia content.

[0172] In the software implementation, the device influence coefficient and shooting device data of the material are first extracted by calling the image processing library or custom algorithm. These data are usually stored in the metadata of the material (such as EXIF ​​data). The software then inputs these data into the optimization model, usually using linear regression, machine learning algorithms or custom adjustment functions to calculate the tone adjustment parameters. After calculating the parameters, the software modifies the material through image processing functions (such as adjusting brightness and color temperature). After the modification is completed, the software recombines the adjusted material with other unlabeled materials and generates the final multimedia content. This entire process can be implemented by writing automated scripts or using existing image processing tools to ensure efficient and consistent processing.

[0173] The color consistency is verified based on the generated content. If a color inconsistency is detected, it is readjusted until the consistency reaches the preset standard.

[0174] After multimedia content is generated, the system needs to verify its color consistency to ensure that all materials are consistent in terms of brightness and color temperature. The verification process first analyzes the overall color tone of the generated content, calculates the comprehensive color tone parameters of the content, and compares them with the preset color consistency standards. If the color tone parameters are found to exceed the preset standards, the system will identify the specific materials that cause the color inconsistency and automatically trigger the adjustment process to reapply the color adjustment parameters to these materials, combine them again after adjustment to generate content, and repeat the verification process until the generated multimedia content meets the preset color consistency standards.

[0175] The software first uses an image processing algorithm to extract the overall tonal parameters of the generated content, including average brightness, contrast, color temperature, etc. The software then compares these parameters with the preset tonal consistency standards and determines whether the content meets the standards based on the set threshold. If an inconsistency is detected, the software automatically analyzes each material in the generated content, identifies materials with large tonal deviations, and corrects them using the previously calculated tonal adjustment parameters. Once the correction is complete, the software combines the generated content again and repeats the above verification and adjustment steps. The entire process is implemented through the cooperation of automated scripts and image processing libraries to ensure that the processing flow is efficient and accurate until the generated content meets the preset standards.

[0176] like Figure 2 The multimedia content intelligent generation system based on a massive material library shown includes a material acquisition and data integration module, a lighting analysis and color tone evaluation module, a color tone consistency recognition and marking module, an optimization model application and material reorganization module, and a color tone consistency verification and dynamic adjustment module;

[0177] The material acquisition and data integration module obtains multiple materials to be combined from the massive material library, collects the lighting condition data and shooting equipment data of each material, and transmits these data to the central processing system for storage and monitoring;

[0178] The lighting analysis and color tone evaluation module analyzes the lighting condition data transmitted to the central processing system to evaluate whether color tone inconsistencies will occur during the material combination process;

[0179] The color consistency recognition and marking module further analyzes the collected shooting equipment data, identifies the materials with inconsistent colors and marks them when it is assessed that there are color inconsistencies in the material combination process;

[0180] The optimization model application and material reorganization module adjusts the materials marked as inconsistent in tone by applying the optimization model and recombines them to generate multimedia content;

[0181] The tone consistency verification and dynamic adjustment module verifies the tone consistency based on the generated content. If tone inconsistency is detected, it is readjusted until the consistency reaches the preset standard.

[0182] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0183] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center containing one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0184] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0185] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0186] In the several embodiments provided in the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0187] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0188] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0189] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for intelligently generating multimedia content based on a massive material library, characterized in that: The specific steps include: Obtain multiple materials to be combined from the massive material library, collect the lighting condition data and shooting equipment data of each material, and transmit these data to the central processing system for storage and monitoring; Analyze the lighting condition data transmitted to the central processing system to evaluate whether there will be any color inconsistencies during the material combination process; The specific steps include: Extract the lighting condition data of each material transmitted to the central processing system, including three types of data: average brightness, exposure value and color temperature; Pre-process the three types of data, namely, average brightness, exposure value and color temperature of each extracted material; Analyze the average brightness, exposure value and color temperature of each pre-processed material to generate the lighting consistency coefficient and color temperature balance index respectively; The logic for obtaining the illumination consistency coefficient and the color temperature balance index is as follows: The average brightness, exposure value and color temperature of each preprocessed material are extracted, and the average brightness, exposure value and color temperature of each preprocessed material are calibrated as L m 、E m and T m , L m represents the average brightness of the mth material after preprocessing, E m represents the exposure value of the preprocessed m-th material, T m Indicates the color temperature of the mth material after preprocessing, , k is a positive integer; The average brightness L of the i-th material i and the average brightness L of the jth material j For comparison, the value range of i and j is , i and j are both positive integers, and the brightness difference between the i-th material and the j-th material is calculated. The specific calculation formula is as follows: ； In the formula, is the brightness difference between the i-th material and the j-th material; The exposure value E for the i-th material i and the exposure value E of the jth material j Compare and calculate the exposure difference between the Ith material and the Jth material. The specific calculation formula is as follows: ; In the formula, is the exposure difference between the i-th material and the j-th material; Calculate the exposure adjustment factor between the i-th material and the j-th material. The specific calculation formula is as follows: ; In the formula, is the exposure adjustment factor between the i-th material and the j-th material; The brightness difference between the i-th material and the j-th material is corrected by combining the exposure adjustment factor between the i-th material and the j-th material. The specific calculation formula is as follows: ; In the formula, is the brightness difference between the i-th material and the j-th material after correction; Calculate the lighting consistency coefficient. The specific calculation formula is as follows: ; In the formula, C L is the illumination consistency coefficient; The color temperature T of the i-th material i and the color temperature T of the jth material j Compare and calculate the color temperature difference between the i-th material and the j-th material. The specific calculation formula is as follows: ; In the formula, is the color temperature difference between the i-th material and the j-th material; Calculate the smoothing factor between the i-th material and the j-th material. The specific calculation formula is as follows: ; In the formula, is the smoothing factor between the i-th material and the j-th material; The color temperature difference between the i-th material and the j-th material is corrected by combining the smoothing factor between the i-th material and the j-th material. The specific calculation formula is as follows: ; In the formula, is the color temperature difference between the i-th material and the j-th material after correction; Calculate the color temperature balance index. The specific calculation formula is as follows: ; In the formula, I T is the color temperature balance index; The generated illumination consistency coefficient and color temperature balance index are used to construct a color tone consistency evaluation model, generate a color tone consistency coefficient, and compare the generated color tone consistency coefficient with a preset color tone consistency threshold to evaluate whether color tone inconsistency will occur during the material combination process; When it is assessed that there is a problem of inconsistent color tones in the material combination process, further analyze the collected shooting equipment data, identify and mark the materials with inconsistent color tones; Applying an optimization model to adjust materials marked as inconsistent in tone, and recombining to generate multimedia content; The color consistency is verified based on the generated content. If a color inconsistency is detected, it is readjusted until the consistency reaches the preset standard.

2. The method for intelligently generating multimedia content based on a massive material library according to claim 1, characterized in that: The generated lighting consistency coefficient C L and color temperature balance index I T Construct a hue consistency evaluation model, generate a hue consistency coefficient CTCC by weighted summation, and compare the generated hue consistency coefficient CTCC with the preset hue consistency threshold CTCC yz Compare and evaluate whether there will be any color inconsistency during the material combination process. The specific analysis is as follows: like , there will be no color inconsistency during the material combination process; like , there will be color inconsistency problems during the material combination process.

3. The method for intelligently generating multimedia content based on a massive material library according to claim 2, characterized in that: When it is assessed that there is a problem of inconsistent color tone in the material combination process, further analyze the collected shooting device data, identify and mark the materials with inconsistent color tone, which specifically includes the following steps: Acquire the shooting device data associated with each material and perform preprocessing after acquisition; Extracting shutter speed and sensitivity of each material from pre-processed camera data; The shutter speed and sensitivity of each extracted material are standardized, and after the standardization, a comprehensive analysis is performed to generate the equipment impact coefficient of each material by building an equipment impact assessment model; The device impact coefficient of each generated material is compared with a preset device impact threshold, and materials whose device impact coefficient exceeds the preset device impact threshold are identified and marked as materials with inconsistent tones.

4. The method for intelligently generating multimedia content based on a massive material library according to claim 3, characterized in that: The shutter speed and sensitivity of each extracted material are standardized, and after the standardization, a comprehensive analysis is performed to generate the device impact coefficient of each material by building a device impact assessment model, which specifically includes the following steps: The shutter speed and sensitivity of each extracted material are calibrated as V m and ISO m , V m Indicates the shutter speed of the mth material, ISO m Indicates the sensitivity of the mth material, , k is a positive integer; Get the maximum and minimum shutter speeds of all materials, and mark the maximum and minimum shutter speeds of all materials as V max and V min , get the maximum and minimum values ​​of the sensitivity of all materials, and calibrate the maximum and minimum values ​​of the sensitivity of all materials as ISO max and ISO min ; The shutter speed V for each clip m and ISO m Perform standardization, the standardization formula is as follows: ; ; In the formula, is the normalized shutter speed of the mth material, is the normalized sensitivity of the mth material; Collect several shutter speeds, several sensitivities and corresponding equipment influence coefficients of the m-th material after standardization in the past period of time, and calibrate them as , and , x represents the shutter speeds, sensitivities and corresponding equipment influence coefficients of the m-th material after standardization in the past period of time, , h is a positive integer, and the data collected in the past period of time is formed into a historical data set; Select the multivariate regression model as the equipment impact assessment model, and train it through the historical data set to determine the value of the regression coefficient according to the formula: ； In the formula, , and is the regression coefficient; By minimizing the error between the predicted value and the actual value, the regression coefficient is optimized and finally determined. , and The value of Using the final regression coefficient, the constructed device impact assessment model is input with the shutter speed of the mth material generated in real time. and sensitivity , generate the device influence coefficient DIF of the mth material in real time m .

5. The method for intelligently generating multimedia content based on a massive material library according to claim 4, characterized in that: The device influence coefficient DIF of each material generated m With the pre-set device impact threshold DIF yz Compare and identify the equipment impact coefficient DIF m Exceeds the preset device impact threshold DIF yz The material is marked as inconsistent in tone. The specific analysis is as follows: like , do not mark the material as tonally inconsistent material; like , marking the clip as tonally inconsistent.

6. The method for intelligently generating multimedia content based on a massive material library according to claim 5, characterized in that: The optimization model is applied to adjust the materials marked as inconsistent in hue, and the materials are recombined to generate multimedia content, which specifically includes the following steps: extracting the equipment influence coefficient and shooting equipment data of the materials marked as inconsistent in hue, inputting these data into the optimization model, and calculating the hue adjustment parameters, including the exposure adjustment factor and the color temperature correction coefficient; applying the hue adjustment parameters, performing image processing on the materials, adjusting their brightness and color temperature, so that they are consistent with the lighting conditions and color temperature of other materials; finally, recombining the adjusted materials with other unmarked materials to generate multimedia content.

7. A multimedia content intelligent generation system based on a massive material library, used to implement the multimedia content intelligent generation method based on a massive material library as described in any one of claims 1 to 6, characterized in that: It includes material collection and data integration module, lighting analysis and color tone evaluation module, color tone consistency recognition and marking module, optimization model application and material reorganization module and color tone consistency verification and dynamic adjustment module; The material acquisition and data integration module obtains multiple materials to be combined from the massive material library, collects the lighting condition data and shooting equipment data of each material, and transmits these data to the central processing system for storage and monitoring; The lighting analysis and color tone evaluation module analyzes the lighting condition data transmitted to the central processing system to evaluate whether color tone inconsistencies will occur during the material combination process; The color consistency recognition and marking module further analyzes the collected shooting equipment data, identifies the materials with inconsistent colors and marks them when it is assessed that there are color inconsistencies in the material combination process; The optimization model application and material reorganization module adjusts the materials marked as inconsistent in tone by applying the optimization model and recombines them to generate multimedia content; The tone consistency verification and dynamic adjustment module verifies the tone consistency based on the generated content. If tone inconsistency is detected, it is readjusted until the consistency reaches the preset standard.

Images