Dynamic special effect processing method and device, storage medium and related equipment

Abstract

The invention discloses a dynamic special effect processing method and device, a storage medium and related equipment, which are applied to the technical field of dynamic special effect rendering, and are used for performing plot analysis on an obtained to-be-processed video to obtain a video plot background, an emotional tone in the plot and a key plot node, and performing dynamic special effect processing according to the video plot background and the emotional tone in the plot. The method comprises the following steps: respectively performing emotion reasoning and emotion trajectory construction to obtain an emotion label and an emotion trajectory of a role, identifying a plot turning point through the emotion trajectory of the role and a key plot node, and obtaining a bullet screen analysis result corresponding to a video to be processed according to the emotion label, the obtained emotion intensity, the emotion association role, the plot turning point and the to-be-processed video. And determining a triggering opportunity, a triggering emotion and a triggering special effect, and performing emotion special effect rendering through the triggering opportunity, the triggering emotion and the triggering special effect.

Description

Technical Field

[0001] The present application relates to the technical field of dynamic special effect rendering, and more specifically, to a dynamic special effect processing method, device, storage medium and related equipment. Background Art

[0002] With the popularization of Internet video content, video playback platforms pay more and more attention to the immersive experience of users. Among them, emotional special effects, as one of the important means to enhance the user experience, have been widely used.

[0003] The triggering methods of existing emotional special effects usually rely on static emotional tags. Since the generation of emotional tags is usually based on single emotional recognition and lacks a comprehensive analysis of complex plots, the triggering timing of special effects is not accurate enough, and there may be situations of premature or late triggering. Therefore, the triggering methods of emotional special effects relying on static emotional tags have certain limitations. When multiple emotions coexist, the triggering of existing emotional special effects does not fully consider the dynamic transition between emotions, and it is difficult to achieve smooth transition and natural connection of emotional special effects. The triggering of emotional special effects usually does not fully combine user preferences and plot development, and the triggering of special effects is not personalized and accurate enough, resulting in inaccurate triggering timing of emotional special effects.

[0004] Therefore, how to make the triggering timing of emotional special effects accurate to improve the user viewing experience is an urgent problem to be solved in the present application. Summary of the Invention

[0005] In view of this, the present application discloses a dynamic special effect processing method, device, storage medium and related equipment, aiming to achieve accurate triggering and rendering efficiency of emotional special effects, enhance the immersion of the audience, and improve the user viewing experience.

[0006] To achieve the above object, the disclosed technical solutions are as follows:

[0007] The first aspect of the present application discloses a dynamic special effect processing method, and the method includes:

[0008] Perform a plot analysis on the obtained video to be processed to obtain the video plot background, the emotional tone in the plot, and the key plot nodes;

[0009] According to the video plot background and the emotional tone in the plot, perform emotional reasoning and emotional trajectory construction respectively to obtain emotional tags and the emotional trajectory of the character;

[0010] Identify the plot turning points through the emotional trajectory of the character and the key plot nodes;

[0011] Determine the triggering time, triggering emotion, and triggering special effects according to the emotion label, the obtained emotion intensity, the emotion-related role, the plot turning point, and the barrage analysis result corresponding to the video to be processed;

[0012] Perform emotion special effect rendering through the triggering time, the triggering emotion, and the triggering special effects.

[0013] Preferably, the process of obtaining the key plot nodes includes:

[0014] Perform script line analysis, visual retrieval and recognition, character dynamic behavior analysis, and emotional tone analysis on the video to be processed respectively to complete the process of obtaining the key plot nodes.

[0015] Preferably, the performing emotion reasoning and emotional trajectory construction respectively according to the video plot background and the emotional tone in the plot to obtain the emotion label and the emotional trajectory of the role includes:

[0016] Obtain the emotional changes of the role in the video to be processed at each time point;

[0017] Extract the emotion label according to the video plot background, the emotional tone in the plot, and the emotional changes of the role at each time point;

[0018] Based on the emotional tone in the plot, determine the emotional fluctuations between time points, and infer the emotional trajectory of the role according to the emotional fluctuations between time points.

[0019] Preferably, the based on the emotional tone in the plot, determining the emotional fluctuations between time points, and inferring the emotional trajectory of the role according to the emotional fluctuations between time points includes:

[0020] For each frame of the video to be processed, extract the emotion intensity data of the same role from the emotion report corresponding to each frame of the video; the emotion intensity data represents the emotional fluctuations of the same role between different time points;

[0021] Generate a smooth curve of emotion data through a preset interpolation algorithm and the emotion intensity data of the same role;

[0022] Establish an emotional trajectory curve according to the smooth curve of emotion data and the time point data of the video to be processed obtained in advance;

[0023] Perform emotion fluctuation recognition on the emotional trajectory curve to obtain the preset fluctuation moments and their emotion fluctuation data;

[0024] Infer the emotional trajectory of the role according to the preset fluctuation moments and their emotion fluctuation data.

[0025] Preferably, identifying a plot turning point through the emotional trajectory of the character and the key plot nodes includes:

[0026] Performing emotional intensity identification based on the emotional trajectory of the character and the key plot nodes to obtain an identification result;

[0027] If the identification result indicates that the increase value of the emotional intensity within a preset time interval is greater than or equal to a preset value, determine the time interval corresponding to the identification result as the plot turning point.

[0028] Preferably, performing emotional special effect rendering through the triggering timing, the triggering emotion, and the triggering special effect includes:

[0029] Determine the triggering moment and the termination moment of the emotional special effect rendering according to the triggering timing;

[0030] Calculate the time difference between the triggering moment and the termination moment;

[0031] Determine the time difference as the rendering duration;

[0032] Perform emotional special effect rendering through the rendering duration, the triggering emotion, and the triggering special effect.

[0033] Preferably, it further includes:

[0034] Respond to user role preferences;

[0035] Determine the triggering timing and the triggered emotional special effect according to the user role preferences, the plot turning point, the emotional trajectory of the character, and the bullet screen analysis result corresponding to the video to be processed.

[0036] The second aspect of the present application discloses a dynamic special effect processing device, and the device includes:

[0037] An analysis unit for performing plot analysis on the acquired video to be processed to obtain the video plot background, the emotional tone in the plot, and the key plot nodes;

[0038] An inference construction unit for respectively performing emotion inference and emotional trajectory construction according to the video plot background and the emotional tone in the plot to obtain an emotion label and the emotional trajectory of the character;

[0039] An identification unit for identifying a plot turning point through the emotional trajectory of the character and the key plot nodes;

[0040] A determination unit for determining the triggering timing, the triggering emotion, and the triggering special effect according to the emotion label, the acquired emotional intensity, the emotion-related character, the plot turning point, and the bullet screen analysis result corresponding to the video to be processed.

[0041] A rendering unit for performing mood special effect rendering based on the triggering timing, the triggering mood, and the triggering special effect.

[0042] A third aspect of the present application discloses a storage medium, which includes stored instructions. When the instructions run, the device where the storage medium is located is controlled to execute the dynamic special effect processing method according to any one of the first aspects.

[0043] A fourth aspect of the present application discloses an electronic device, which includes a memory and one or more instructions. One or more instructions are stored in the memory and are configured to be executed by one or more processors to execute the dynamic special effect processing method according to any one of the first aspects.

[0044] As can be seen from the above technical solutions, the present application discloses a dynamic special effect processing method, device, storage medium, and related equipment. The method analyzes the plot of the video to be processed, obtains the video plot background, the emotional tone in the plot, and the key plot nodes. According to the video plot background and the emotional tone in the plot, emotional reasoning and emotional trajectory construction are respectively performed to obtain the emotional label and the emotional trajectory of the character. The plot turning point is identified through the emotional trajectory of the character and the key plot nodes. According to the emotional label, the obtained emotional intensity, the emotion-related character, the plot turning point, and the bullet screen analysis result corresponding to the video to be processed, the triggering timing, the triggering mood, and the triggering special effect are determined. Mood special effect rendering is performed through the triggering timing, the triggering mood, and the triggering special effect.

[0045] Through the above solutions, in-depth plot analysis, emotional reasoning, and emotional trajectory construction of the video to be processed are carried out to accurately analyze the video plot and emotional fluctuations. The triggering timing, the triggering mood, and the triggering special effect are determined according to the obtained emotional intensity, the emotion-related character, the plot turning point, and the bullet screen analysis result corresponding to the video to be processed. Mood special effect rendering is performed through the triggering timing, the triggering mood, and the triggering special effect to achieve natural rendering. The rendering duration and intensity of the special effect are controlled according to the emotional intensity curve for mood special effect rendering to ensure that the special effect display is consistent with the plot development. At the same time, when multiple emotions coexist, mood special effect rendering is performed through the triggering timing, the triggering mood, and the triggering special effect, and multiple emotion special effects can be smoothly transitioned, alternated, and naturally connected to avoid visual conflicts, realizing accurate triggering and rendering efficiency of emotion special effects. In addition, a personalized special effect experience is provided according to the user's emotional preference, effectively enhancing the audience's immersion and improving the user's viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the provided drawings.

[0047] Figure 1 It is a schematic flowchart of a dynamic special effect processing method disclosed in an embodiment of the present application;

[0048] Figure 2 It is a schematic flowchart of another dynamic special effect processing method disclosed in an embodiment of the present application;

[0049] Figure 3 It is a schematic structural diagram of a dynamic special effect processing device disclosed in an embodiment of the present application;

[0050] Figure 4 It is a schematic structural diagram of an electronic device disclosed in an embodiment of the present application. Detailed implementation manners

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0052] In the present application, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, the element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0053] As can be seen from the background art, the triggering methods of existing emotion special effects usually rely on static emotion tags. Since the generation of emotion tags is usually based on single emotion recognition and lacks a comprehensive analysis of complex plots, the triggering timing of special effects is not accurate enough, and the situation of premature or late triggering may occur. Therefore, the triggering method of emotion special effects relying on static emotion tags has certain limitations. When multiple emotions coexist, the triggering of existing emotion special effects does not fully consider the dynamic transition between emotions, and it is difficult to achieve smooth transition and natural connection of emotion special effects. The triggering of emotion special effects usually does not fully combine user preferences and plot development, and the triggering of special effects appears to be insufficiently personalized and accurate, resulting in inaccurate triggering timing of emotion special effects. Therefore, how to make the triggering timing of emotion special effects accurate to improve the user viewing experience is an urgent problem to be solved in this application.

[0054] A dynamic special effect processing method, device, storage medium and related equipment provided by this application can be used in the field of dynamic special effect rendering technology or other fields. Other fields refer to any fields other than the field of dynamic special effect rendering technology. For example, the field of affective computing, the field of video content analysis, etc. The above are only examples and do not limit the application fields of a dynamic special effect processing method, device, storage medium and related equipment provided by this application.

[0055] To solve the above problems, this application discloses a dynamic special effect processing method, device, storage medium and related equipment, which conducts in-depth plot analysis, emotion reasoning and emotional trajectory construction on the video to be processed, so as to accurately analyze the video plot and emotional fluctuations, obtain the emotional intensity, emotion-related characters, plot turning points and the barrage analysis results corresponding to the video to be processed, determine the triggering timing, triggering emotion and triggering special effects, and through the triggering timing, triggering emotion and triggering special effects, perform emotion special effect rendering to achieve natural rendering, and perform emotion special effect rendering that controls the rendering duration and intensity of the special effects according to the emotional intensity curve to ensure that the special effect display is consistent with the plot development. At the same time, through the triggering timing, triggering emotion and triggering special effects for emotion special effect rendering, when multiple emotions coexist, it is possible to smoothly transition and alternate multiple emotion special effects and make natural connections, avoid visual conflicts, and achieve accurate triggering and rendering efficiency of emotion special effects. In addition, providing a personalized special effect experience according to the user's emotion preferences effectively enhances the audience's immersion and improves the user viewing experience. The specific implementation method is specifically described through the following embodiments.

[0056] Refer to Figure 1 As shown, a dynamic special effect processing method disclosed in an embodiment of this application is applicable to this dynamic special effect processing system. The dynamic special effect processing system includes a plot analysis module, an emotion analysis module, an emotion special effect triggering and timing decision module, and an emotion characteristic rendering and optimization module. The dynamic special effect processing method mainly includes the following steps:

[0057] S101: Analyze the plot of the video to be processed, and obtain the video plot background, the emotional tone in the plot, and the key plot nodes.

[0058] Through the plot analysis module, analyze the video to be processed for the video plot background and emotional tone, and obtain the video plot background, the emotional tone in the plot, and the key plot nodes.

[0059] The core task of the plot analysis module is to understand the overall plot of the video to be processed, analyze the emotional tone and plot trend therein, and provide data support for subsequent emotion analysis and emotion label extraction. By deeply analyzing the video content, generate the video plot background to provide an analysis basis for the emotion reasoning module.

[0060] Plot background analysis:

[0061] First, analyze the overall plot of the video to be processed, and identify the main plot developments in the plot. For example, the video may contain the growth story of a character, the generation and resolution of conflicts, emotional climaxes and ups and downs, etc. Through in-depth analysis of the script, scene changes, character behaviors, dialogue content, etc., the system can understand the emotional tone of the plot, such as from calm to conflict, and then to reconciliation.

[0062] Judgment of emotional tone:

[0063] Through the plot analysis of the video, the dynamic special effect processing system can determine the emotional tendency at each stage (for example, sadness, anger, anxiety, joy, etc.). For example, after character A loses a loved one, the emotional tone of the plot may be sadness, and then the plot development may transition to anger, and finally the emotion tends to be relieved.

[0064] Marking of key plot nodes:

[0065] The dynamic special effect processing system identifies the key plot nodes in the plot. These key plot nodes are usually the turning points of emotional fluctuations, such as the outbreak of emotional conflicts or the liberation of character emotions. Through the analysis of the key plot nodes, the dynamic special effect processing system can provide marks of time points and plot changes for subsequent emotion analysis.

[0066] It should be noted that in the dynamic special effect processing system, the key plot nodes are the main structural events in the plot, such as the beginning, climax or end of the story, etc. These are identified by the plot analysis module through multi-modal information processing technologies such as script, visual and audio analysis (including lines, scene switching, facial expressions, character behaviors, etc.). The structural nodes (such as the beginning, climax and end of the story) identified by the plot analysis module through script, visual and audio analysis are the basic components of the plot, which determine the context of the plot development, but do not directly trigger emotion special effects.

[0067] The process by which the specific dynamic special effect processing system identifies key plot nodes in the plot is as follows:

[0068] The dynamic special effect processing system performs script line analysis, visual retrieval recognition, character dynamic behavior analysis, and emotional tone analysis on the video to be processed respectively to complete the process of obtaining key plot nodes.

[0069] Script line analysis:

[0070] Using natural language processing technology, analyze the keywords and sentences in the script, especially those lines with strong emotions (e.g., "You will never forgive me"). These emotional statements usually indicate the occurrence or progression of key plots.

[0071] Visual clue recognition:

[0072] Analyze the visual signals in the video, including changes in the scene (such as the background darkening, transition effects, or scene switching) and changes in the facial expressions and body movements of the characters (such as from tense to relaxed, from smiling to crying, etc.). These changes usually occur at plot twists or climaxes.

[0073] Character behavior pattern:

[0074] Analyze the dynamic behavior of the characters to identify whether there are obvious changes in their behaviors. For example, sudden changes in character behavior or sudden pauses in behavior may imply the emergence of key plots.

[0075] Emotional tone analysis: Combine emotional tone detection to analyze the change in emotional color in the lines, so as to judge whether the key plot has entered the climax, the end, or the moment when the core conflict is about to occur.

[0076] S102: According to the video plot background and the emotional tone in the plot, perform emotion reasoning and emotional trajectory construction respectively to obtain emotion labels and the emotional trajectory of the character.

[0077] In S102, based on the video plot background and the emotional tone in the plot provided by the emotion analysis module, perform character emotion reasoning and emotion label extraction, and speculate on possible plot turning points.

[0078] A plot turning point is a moment of significant change in the plot or emotion, usually manifested as a drastic emotional fluctuation or mutation. The recognition of turning points is completed by the emotion analysis module. By analyzing the emotional fluctuations of the characters, the turning point moment is accurately located when there is a sudden change in emotional intensity or a sharp change in emotion.

[0079] The recognition of specific plot turning points is as follows:

[0080] Based on the emotional trajectory of the characters, the emotion analysis module identifies the moments of emotional outbursts, which often correspond to the climax or emotional turning points of the plot. For example, when Character A has an emotional outburst during a conflict with other characters, the emotional intensity may suddenly increase from 50% to 80%, and this is a plot turning point of emotional outburst.

[0081] The prediction process of the turning point is as follows:

[0082] The emotion analysis module not only identifies the current turning points, but also can speculate on future emotional changes according to the plot development trend. For example, Character B may have an outburst of anxiety in the upcoming emotional conflict, and this possible turning point is marked in advance to help the dynamic special effects processing system trigger the corresponding emotional special effects at the correct moment.

[0083] The specific process of obtaining the emotion labels and the emotional trajectory of the characters is shown in A1 - A3.

[0084] A1: Obtain the emotional changes of the characters in the video to be processed at each time point.

[0085] To facilitate the understanding of the emotional changes of the characters at each time point, an example is given here for illustration:

[0086] For example, Character A shows anger at 14:30:28 and a transition to relief at 14:30:37. That is, the emotional changes of Character A at each time point are from 14:30:28 to 14:30:37.

[0087] A2: Extract emotion labels based on the video plot background, the emotional tone in the plot, and the emotional changes of the characters at each time point.

[0088] In A2, the emotion analysis module extracts emotion labels according to the video plot background, the emotional tone in the plot, and the emotional changes of the characters at each time point. For example, Character A shows anger (intensity 70%) at a certain time point (such as 14:30:26), and according to the progress of the plot background, the emotion labels will be updated dynamically. For example, it may show a transition from anger to relief at the next few time points (such as anger at 14:30:28 and relief at 14:30:46).

[0089] The emotional tone in the plot can be expressed as: for example, from calm to conflict, and then to reconciliation.

[0090] A3: Based on the emotional tone in the plot, determine the emotional fluctuations between time points, and infer the emotional trajectory of the characters according to the emotional fluctuations between time points.

[0091] In A3, the emotion analysis module infers the emotional trajectory of a character based on the emotional tone generated by the plot analysis. For example, the emotion of character A may start from sadness, gradually turn to anger, and finally tend to relief. In this process, the emotion analysis module will generate a dynamic emotional trajectory through the emotional fluctuations between time points.

[0092] It should be noted that the emotional trajectory describes the emotional change path of a character in a video. This emotional change path is composed of emotional data at multiple time points, and through the changes of these emotional data at multiple time points, an emotional fluctuation trajectory is formed. The core idea is to generate a continuous emotional trajectory through the emotional fluctuations between time points, and the changes in emotional fluctuations can be caused by certain characteristics of emotional mutations or drastic fluctuations.

[0093] Specifically, based on the emotional tone in the plot, the emotional fluctuations between time points are determined, and the process of inferring the emotional trajectory of a character according to the emotional fluctuations between time points is shown in B1 - B5.

[0094] B1: For each frame of the video to be processed, extract the emotional intensity data of the same character from the emotional report corresponding to each frame of the video; the emotional intensity data represents the emotional fluctuations of the same character between different time points.

[0095] Among them, the emotional intensity data is the time - point emotion.

[0096] Extract the emotional intensity data of the character from the emotional report corresponding to each frame or a segment of the video. For example: At time point T1, the emotion of character A is sadness (intensity 50%). At time point T2: The emotion of character A is sadness (intensity 60%). At time point T3: The emotion of character A is relief (intensity 40%).

[0097] B2: Generate a smooth curve of emotional data through a preset interpolation algorithm and the emotional intensity data of the same character.

[0098] Among them, the preset interpolation algorithms include but are not limited to spline interpolation, linear interpolation, etc.

[0099] In order to remove noise and represent the emotional trajectory more smoothly, a preset interpolation algorithm (such as spline interpolation or linear interpolation) can be used to generate emotional data between two time points. For example, if the emotional intensity at T1 is 30% and at T2 is 45%, the emotional values at any moment between T1 and T2 (such as T1.5, T1.7, etc.) can be estimated through interpolation to obtain their emotional intensities and form a smooth curve.

[0100] B3: Establish an emotional trajectory curve based on the smooth curve of emotional data and the time - point data of the video to be processed obtained in advance.

[0101] In B3, based on the time-point data and interpolation results (i.e., the smoothed curve of emotion data) of the video to be processed obtained, an emotional trajectory curve of the character is established, and this curve reflects the emotional changes of the character in the plot.

[0102] B4: Identify emotional fluctuations in the emotional trajectory curve to obtain the preset fluctuation moments and their emotional fluctuation data.

[0103] The preset fluctuation moments are the moments of intense fluctuations. For example, the moments when the emotion intensity rapidly increases from low to high, or the moments when the emotional fluctuation range is greater than the preset fluctuation range. The preset fluctuation range is set according to the actual situation, and this application does not make specific limitations.

[0104] The specific process of emotional fluctuation identification is as follows:

[0105] Analyze the generated emotional trajectory curve above to identify the moments of intense fluctuations. For example, the moments when the emotion intensity rapidly increases from low to high, or the relatively large moments when the emotional fluctuation range is greater than the preset fluctuation range. These fluctuation points usually represent the moments of drastic emotional changes, which may imply the turning points or climax parts of the plot. For example, if at a certain time point during the plot development, the emotion of the character changes from very stable (low intensity) to intense emotion (high intensity), this moment can be regarded as the turning point of emotional fluctuation.

[0106] B5: Infer the emotional trajectory of the character based on the preset fluctuation moments and their emotional fluctuation data.

[0107] Through the above process, a dynamic emotional trajectory of the character can be inferred from the discrete time points and emotional fluctuation data, and this emotional trajectory accurately captures the emotional ups and downs of the character as the plot develops.

[0108] S103: Identify the plot turning points through the emotional trajectory of the character and the key plot nodes.

[0109] In S103, based on the emotional trajectory of the character and the key plot nodes, perform emotion intensity identification to obtain the identification result. If the identification result indicates that the increase value of the emotion intensity within the preset time interval is greater than or equal to the preset value (for example, when the character A has an emotional outburst during the conflict with other characters, the emotion intensity may suddenly increase from 50% to 80%, and this is a turning point of emotional outburst at this time), determine the time interval corresponding to the identification result as the plot turning point.

[0110] Among them, the preset value can be 75%, 80%, etc. The determination of the preset value is set according to the actual situation, and this application does not make specific limitations.

[0111] The emotion analysis module generates an emotion report based on the emotion label, emotion intensity, emotion-related roles, and possible plot turning points at each time point. The emotion report will be passed to the emotion special effect triggering and timing decision module for triggering special effect decisions.

[0112] S104: Determine the triggering time, triggering emotion and triggering special effect according to the emotion label, the acquired emotion intensity, the emotion-related role, the plot turning point and the bullet screen analysis result corresponding to the video to be processed.

[0113] The core task of the emotional special effects triggering and timing decision module is to determine the triggering timing, select the triggering emotion, and estimate the starting and ending intervals of the emotion based on the intensity of the emotion, the role associated with the emotion, the plot turning point, the user role preference, the trend of emotional fluctuations, and the barrage analysis results corresponding to the video to be processed. The emotional special effects triggering and timing decision module is responsible for performing the following three main tasks:

[0114] (1) Determine the triggering time:

[0115] The judgment of the triggering timing determines whether to enter the decision-making stage of emotional special effects triggering. The following conditions will trigger the judgment of special effects timing:

[0116] Identifying plot turning points:

[0117] The dynamic special effects processing system identifies key plot nodes in the video, such as emotional conflicts, emotional climaxes, etc., and determines whether it is the time to trigger emotional special effects.

[0118] The specific process of determining whether it is time to trigger an emotional effect is as follows:

[0119] 1. Emotional fluctuation monitoring: The system first monitors the character's emotional changes in real time, especially at moments of dramatic fluctuations. For example, a character may experience an emotional transition from calm to anger or sadness. The system analyzes emotional changes through multimodal data (such as the character's facial expressions, dialogue intonation, behavioral movements, etc.) and identifies whether it is a turning point.

[0120] 2. Emotional conflict and climax analysis: In the plot, emotional conflict (such as intense dialogue or behavioral conflict between characters) or emotional climax (such as a violent outburst of character emotions) are typical turning points. When the system detects these emotional fluctuations, especially when the emotional intensity rises sharply, it will identify these moments as turning points.

[0121] 3. Determine when to trigger special effects: When the system identifies a turning point and the emotional intensity at that moment exceeds the set trigger threshold (for example, the intensity of anger or sadness reaches 70%), the system will determine that this is a time to trigger an emotional special effect.

[0122] For example, in a drama, Character A and Character B have a heated conversation. Character A is emotional and shouts, "How could you do this!" The system detects that at this moment, the intensity of Character A's anger quickly exceeds the set threshold (such as 70%), and the amplitude of emotional fluctuations is very large. At this time, the system will determine that this moment is a turning point of emotional conflict and judge that this is an appropriate time to trigger the anger emotion special effect to enhance the audience's immersion.

[0123] The value of the anger emotion intensity is obtained by analyzing multi-modal data (such as script lines, character facial expressions, speech intonation, etc.) through a deep learning model in the emotion analysis module. Through these models, the system can predict the emotions of characters at each time point, thereby obtaining the intensity values (such as anger, sadness, etc.) of different emotions and their change trends.

[0124] (2) Emotional changes in user-character preferences:

[0125] If the user selects a certain character and the emotional change of this character reaches the set trigger threshold (for example, the anger emotion intensity reaches 70%), the dynamic special effect processing system determines that this is a time to trigger the emotion special effect.

[0126] (3) Analysis of the growth rate of bullet screen emotions:

[0127] If the growth rate of a certain emotion (such as anger, sadness) in the bullet screen is significant, the system regards it as a concentrated manifestation of the audience's emotions, thereby triggering the corresponding emotion special effect.

[0128] (4) Selected emotions to trigger:

[0129] Once the trigger time is confirmed, the dynamic special effect processing system enters the emotion special effect selection stage. This stage determines the actual triggered emotion special effect based on factors such as user preferences, emotion proportions, and bullet screen data. The trigger logic is as follows:

[0130] Triggered by user characters (with priority):

[0131] If the emotional intensity of the character selected by the user reaches the set threshold (the set threshold is set according to the actual situation and is not specifically limited in this application), the dynamic special effect processing system will preferentially trigger the emotional special effects of this character. For example, if the anger emotional intensity of character A is 50%, the dynamic special effect processing system will preferentially trigger the anger special effect. If the user prefers multiple characters, the dynamic special effect processing system will select which character's emotional special effect to trigger according to the maximum proportion in the emotion report. For example: if the sadness proportion of character A is 60% and the anger proportion of character B is 40%, the dynamic special effect processing system will trigger the sadness special effect because the sadness emotion proportion is the largest. If the emotion proportions of multiple characters are equivalent, the dynamic special effect processing system will preferentially select the emotion with the fastest growth rate according to the emotion growth rate. For example, if the anger emotion growth rate of character A is relatively fast, the dynamic special effect processing system will select to trigger the anger special effect.

[0132] Bullet screen trigger: If there is no emotion related to the character preferred by the user at the current time point (that is, the character not selected by the user or the character emotion does not reach the trigger threshold), the system will select the triggered emotional special effect according to the bullet screen data. The system will analyze the emotion proportion on the current screen and trigger the emotional special effect with the largest proportion. For example, if the bullet screen shows that the anger emotion proportion is 60% and the sadness emotion proportion is 30%, the anger special effect will be triggered.

[0133] If none of the above conditions are met, that is, there is no emotion of the character preferred by the user and the bullet screen data cannot trigger the special effect, the system will stop triggering the special effect to avoid the display of meaningless or inconsistent special effects with the plot and emotional fluctuations.

[0134] Optionally, in response to the user's character preference, determine the trigger timing and the triggered emotional special effect according to the user's character preference, plot turning point, emotional trajectory of the character, and the analysis result of the bullet screen corresponding to the video to be processed.

[0135] Among them, the user's character preference is an optional function. The user selects their favorite character when watching the video for the first time and can modify this character at any time during the viewing process. This selection will affect the subsequent emotional special effect trigger decision. If the user likes multiple characters, the dynamic special effect processing system will record and give priority to considering the emotional states of these characters in the subsequent special effect triggers.

[0136] The triggered emotional special effect is the selected trigger emotion, the start and end intervals of the emotion.

[0137] S105: Perform emotional special effect rendering through the trigger timing, trigger emotion, and trigger special effect.

[0138] In S105, the emotion special effect rendering and optimization module determines the trigger moment and the termination moment of the emotion special effect rendering according to the trigger timing, calculates the time difference between the trigger moment and the termination moment, determines the time difference as the rendering duration, and performs emotion special effect rendering through the rendering duration, the triggered emotion, and the triggered special effect.

[0139] The emotion special effect rendering and optimization module is responsible for the rendering, convergence, and transition of emotion special effects, ensuring that the special effects can be smoothly transitioned and naturally displayed, and ensuring that the rendering duration is consistent with the emotion intensity curve.

[0140] The calculation process of the rendering duration is as follows:

[0141] The rendering duration of the emotion special effect is calculated by the time difference between the trigger moment and the moment when the emotion intensity drops to the lowest threshold.

[0142] The moment when the emotion intensity drops to the lowest threshold refers to the termination moment of the emotion special effect rendering. When the emotion special effect is triggered, the special effect will gradually change according to the emotion intensity until the emotion intensity drops to the set lowest threshold. Once the lowest threshold is reached, the special effect rendering process ends, and at this time, the rendering duration is calculated as the time difference between the trigger moment and the end moment.

[0143] The trigger moment is the time point when the emotion special effect starts to be rendered. This time point is the trigger time extracted from the emotion report (for example, the time when a certain character's angry emotion is triggered).

[0144] Termination moment: When the emotion intensity drops to the lowest intensity threshold set by the system, the rendering of the special effect ends. The dynamic special effect processing system finds the time point with an intensity lower than the lowest threshold in the emotion intensity data provided in the emotion report as the termination moment.

[0145] Among them, the lowest threshold is set according to the actual situation, and this application does not make specific limitations.

[0146] The process of performing emotion special effect rendering through the rendering duration, the triggered emotion, and the triggered special effect is as shown in 1 to 7 below:

[0147] 1. Normal rendering of the current special effect:

[0148] If no new emotion special effect is triggered during the display of the current emotion special effect, the dynamic special effect processing system will normally render the current special effect according to the emotion intensity curve until the rendering duration ends.

[0149] 2. Emotion intensity curve:

[0150] Based on the data in the emotion report, the system generates an emotion intensity change curve to describe the change process of the emotion from the intensity at the trigger time (such as 50%) to the lowest threshold (such as 25%).

[0151] 3. Rendering process:

[0152] During rendering, parameters such as the number of particles and the intensity of light effects of special effects will be adjusted according to the change of emotion intensity.

[0153] Specifically during rendering, the process of adjusting parameters such as the number of particles and the intensity of light effects of special effects according to the change of emotion intensity is shown as follows in 1)-3):

[0154] 1) Association between the number of particles and emotion intensity:

[0155] Low emotion intensity: The number of particles is small, the movement is slow, showing mild emotions. Medium emotion intensity: The number of particles increases, the speed accelerates, showing medium-intensity emotional fluctuations. High emotion intensity: The number of particles is large, the speed is extremely fast, generating a strong visual impact.

[0156] 2) Association between the intensity of light effects and emotion intensity:

[0157] Low emotion intensity: The intensity of light effects is relatively low, and the picture is soft. Medium emotion intensity: The intensity of light effects increases, and the light color may change (such as red when angry). High emotion intensity: The intensity of light effects increases significantly, producing a flickering or brightness-increasing effect to strengthen the emotional climax.

[0158] 3) Dynamic adjustment process:

[0159] The system adjusts the number of particles and the intensity of light effects by real-time monitoring of emotion intensity and according to the preset relationship to ensure that the special effects are consistent with the emotional fluctuations. By means of smooth transition, abrupt changes are avoided, making the special effects and emotional fluctuations synchronize naturally.

[0160] During the dynamic adjustment process, the special effect parameters are adjusted in real time according to the emotion intensity. The stronger the emotion, the more intense the special effect performance, thus enhancing the audience's immersion and emotional experience.

[0161] Suppose in a scene, character A is experiencing an emotional change from anger to calm:

[0162] At the beginning (emotion intensity 30%): The number of particles in the picture is small, the background color is relatively calm, the intensity of light effects is low, and the overall feeling is relatively soft.

[0163] Emotional outburst (emotion intensity 70%): The number of particles increases significantly, the intensity of light effects is enhanced, and a rapidly flickering red light effect may be added. The background may become stronger and brighter to present the outburst of anger.

[0164] Emotion stable (emotion intensity 20%): The number of particles decreases, the light effects gradually weaken, and the picture returns to the original soft state.

[0165] 4. Trigger new emotion special effects during the current special effect display (i.e., convergence logic):

[0166] If a new emotional special effect is triggered during the current special effect display, the dynamic special effect processing system will execute the convergence logic and smoothly transition to the new emotional special effect.

[0167] 5. Convergence logic:

[0168] If a new emotional special effect is triggered, the dynamic special effect processing system will quickly reduce the intensity of the current emotional special effect according to the set minimum intensity threshold (e.g., 25%) until the current emotion reaches the minimum threshold. This process usually occurs within 1 - 2 seconds to avoid the current emotional special effect having too much impact on the display of the new emotion.

[0169] 6. Activation of the new emotional special effect:

[0170] While converging the current emotional special effect, the dynamic special effect processing system activates the new emotional special effect (such as sadness). The dynamic special effect processing system adjusts the number of particles, light effect intensity, etc. of the special effect according to the intensity curve of the new emotion to ensure that the new emotional special effect can be smoothly transitioned and displayed.

[0171] For example, if the currently rendered emotion is anger with an intensity of 50%, and the sadness emotion is triggered, the system will quickly converge the anger special effect within 1 second, while preloading the sadness special effect and starting to smoothly display it.

[0172] 7. Progressive transition of the special effect:

[0173] When the dynamic special effect processing system detects that a new emotional special effect is about to be triggered, or the intensity of the emotional special effect changes, the rendering of the special effect will undergo a progressive transition.

[0174] Transition of the particle special effect: The particle special effect of the current emotion will gradually transition to the particle effect of the new emotion. For example, the number and speed of the particles of the current emotion gradually decrease, while the particles of the new emotion gradually increase.

[0175] Transition of the dynamic light effect: The light effect area of the current emotion will gradually expand to the area of the new emotion, and the transition speed matches the growth rate of the emotion to ensure a smooth transition of the visual effect.

[0176] Transition of the color change: The hue of the current emotion will gradually transition to the hue of the new emotion, and through the adjustment of color saturation and brightness, ensure a natural transition of the visual effect.

[0177] For the convenience of understanding the process of dynamic special effect processing, combined with Figure 2 the following is an example for illustration:

[0178] For example, implementation scenario: Triggering and rendering of emotional special effects in short drama videos;

[0179] Suppose in a short drama video, the user watches a scene of emotional conflict. Character A shows anger due to being hit, while Character B shows sadness due to disappointment.

[0180] Step 1: The plot analysis module analyzes the plot of the short drama video.

[0181] The dynamic special effect processing system analyzes the video content of the video to be processed, and identifies the emotional tone and key plot nodes in the plot. For example, the emotion of Character A changes from sadness to anger, while the emotion of Character B changes from anger to sadness.

[0182] Step 2: The emotion analysis module analyzes the emotions of the characters in the short drama video.

[0183] The dynamic special effect processing system extracts the emotion tags of Character A and Character B at each time point. For example, at 00:05, the anger emotion intensity of Character A is 50%, and the sadness emotion intensity of Character B is 30%.

[0184] Step 3: Generate an emotion report.

[0185] The emotion analysis module generates an emotion report, recording the emotion tags, emotion intensities, and associated characters at each time point.

[0186] Step 4: User character preference selection.

[0187] The user selects Character A as their preferred character during the first viewing. This preference will be given priority in subsequent special effect triggers.

[0188] Step 5: Emotion special effect trigger and decision.

[0189] The dynamic special effect processing system determines the trigger timing based on the data in the emotion report. For example, when the anger emotion intensity of the user's favorite character A reaches the threshold (50%), the system triggers the anger emotion special effect.

[0190] Step 6: Special effect rendering.

[0191] The dynamic special effect processing system calculates the rendering duration of the special effect based on the emotion intensity and the minimum threshold. For example, when the anger emotion drops from 50% to 25%, the rendering duration is 10 seconds. The system generates an emotion intensity change curve to ensure that the special effect intensity is synchronized with the emotion intensity. The number of particles, light effect intensity, etc. of the special effect are adjusted as the emotion intensity changes.

[0192] Step 7: New emotion trigger and convergence.

[0193] If a new emotion (such as a sad emotion) is triggered during rendering, the dynamic special effect processing system will execute the convergence logic: gradually weaken the intensity of the current emotion special effect until it reaches the lowest threshold. During the convergence process, the system activates a new emotion special effect (such as sadness) and ensures a smooth transition.

[0194] Through this application, users can enjoy more natural and immersive visual effects, and the triggering and rendering of emotion special effects can accurately match the video plot, emotional fluctuations, and user preferences.

[0195] The main purpose of this application is to provide a method for triggering and rendering emotion special effects based on video plot analysis and emotion reasoning. The aim is to analyze the emotional trajectory, mood swings, and plot turning points of the characters in the video, trigger emotion special effects at the appropriate time, and precisely control the rendering duration and intensity of the special effects, making the special effect rendering smoother and more natural, and enhancing the user's immersion and viewing experience.

[0196] Innovative or improved parts of this application:

[0197] (1) Extraction of emotion tags based on plot and emotion analysis: Through in-depth analysis of video content, combined with the plot trend, character behavior, and emotional tone, emotion tags are dynamically generated, and the changes and turning points of emotions are inferred based on the emotional trajectory of the characters.

[0198] (2) Judgment of the triggering time of emotion special effects: According to the plot turning points, the emotional changes of the characters preferred by the user, and the emotional growth rate of the bullet comments, judge when to trigger the emotion special effects and select appropriate emotions for rendering.

[0199] (3) Calculation of the rendering duration of emotion special effects: According to the change in emotion intensity and the set lowest threshold, calculate the rendering duration of the special effect, and ensure that the special effect is gradually rendered according to the emotion intensity curve to avoid the abrupt display of the special effect.

[0200] This application of the present invention belongs to the fields of affective computing, video content analysis, and dynamic special effect rendering, etc., and specifically relates to a method for triggering video special effects based on emotion reasoning and plot analysis. This technology can be widely applied to various scenarios such as video playback platforms and on-demand systems to enhance the user experience. By accurately analyzing the video plot and emotional fluctuations, the accurate triggering and natural rendering of emotion special effects are achieved. The rendering duration and intensity of the special effects are controlled according to the emotion intensity curve to ensure that the special effect display is consistent with the plot development. At the same time, this application can smoothly transition and alternate multiple emotion special effects, avoid visual conflicts, and provide a personalized special effect experience according to the user's emotion preferences. These technologies effectively enhance the audience's immersion and viewing experience, and optimize the management and rendering efficiency of system resources, reduce costs, and improve production efficiency.

[0201] Advantages of the embodiments of the present application: Conduct in-depth plot analysis, emotion reasoning, and emotional trajectory construction on the video to be processed, accurately analyze the video plot and emotional fluctuations, obtain the emotional intensity, emotion-related characters, plot turning points, and barrage analysis results corresponding to the video to be processed, determine the triggering timing, triggering emotion, and triggering special effects, and through the triggering timing, triggering emotion, and triggering special effects, perform emotional special effect rendering to achieve natural rendering. Control the rendering duration and intensity of the special effects according to the emotional intensity curve for emotional special effect rendering to ensure that the special effect display is consistent with the plot development. At the same time, through the triggering timing, triggering emotion, and triggering special effects for emotional special effect rendering, when multiple emotions coexist, it can smoothly transition, alternate multiple emotional special effects, and connect naturally, avoiding visual conflicts, and achieving accurate triggering and rendering efficiency of emotional special effects. In addition, provide a personalized special effect experience according to the user's emotional preferences, effectively enhancing the audience's immersion and improving the user's viewing experience.

[0202] Based on the above embodiments Figure 1 A disclosed dynamic special effect processing method, the embodiments of the present application also correspondingly disclose a dynamic special effect processing device, as Figure 3 shown, the dynamic special effect processing device includes:

[0203] An analysis unit 301, configured to perform plot analysis on the obtained video to be processed to obtain the video plot background, the emotional tone in the plot, and the key plot nodes;

[0204] A reasoning and construction unit 302, configured to perform emotion reasoning and emotional trajectory construction respectively according to the video plot background and the emotional tone in the plot to obtain the emotion label and the emotional trajectory of the character;

[0205] An identification unit 303, configured to identify the plot turning point through the emotional trajectory of the character and the key plot nodes;

[0206] A determination unit 304, configured to determine the triggering timing, triggering emotion, and triggering special effects according to the emotion label, the obtained emotional intensity, emotion-related characters, plot turning points, and barrage analysis results corresponding to the video to be processed;

[0207] A rendering unit 305, configured to perform emotional special effect rendering through the triggering timing, triggering emotion, and triggering special effects.

[0208] Further, the analysis unit 301 for obtaining the key plot nodes is specifically configured to perform script line analysis, visual retrieval and identification, human dynamic behavior analysis, and emotional tone analysis on the video to be processed respectively to complete the process of obtaining the key plot nodes.

[0209] Further, the reasoning and construction unit 302 includes:

[0210] An acquisition module for acquiring the emotional changes of a character in a video to be processed at each time point;

[0211] An extraction module for extracting emotion tags according to the video plot background, the emotional tone in the plot, and the emotional changes of the character at each time point;

[0212] A first determination module for determining the emotional fluctuations between time points based on the emotional tone in the plot, and inferring the emotional trajectory of the character according to the emotional fluctuations between time points.

[0213] Furthermore, the first determination module includes:

[0214] An extraction sub-module for extracting the emotional intensity data of the same character from the emotion reports corresponding to each frame of the video to be processed for each frame of the video; the emotional intensity data represents the emotional fluctuations of the same character between different time points;

[0215] A generation sub-module for generating a smooth curve of emotional data through a preset interpolation algorithm and the emotional intensity data of the same character;

[0216] A establishment sub-module for establishing an emotional trajectory curve according to the smooth curve of emotional data and the time point data of the video to be processed obtained in advance;

[0217] An identification sub-module for identifying emotional fluctuations in the emotional trajectory curve to obtain preset fluctuation moments and their emotional fluctuation data;

[0218] An inference sub-module for inferring the emotional trajectory of the character according to the preset fluctuation moments and their emotional fluctuation data.

[0219] Furthermore, the identification unit 303 includes:

[0220] An identification module for performing emotional intensity identification according to the emotional trajectory of the character and the key plot nodes to obtain an identification result;

[0221] A second determination module for, if the identification result indicates that the increase value of the emotional intensity within a preset time interval is greater than or equal to a preset value, determining the time interval corresponding to the identification result as a plot turning point.

[0222] Furthermore, the rendering unit 305 includes:

[0223] A third determination module for determining the trigger moment and the termination moment of the emotional special effect rendering according to the trigger timing;

[0224] A calculation module for calculating the time difference between the trigger moment and the termination moment;

[0225] A fourth determination module, configured to determine the time difference as the rendering duration;

[0226] A rendering module, configured to perform emotional special effect rendering based on the rendering duration, the triggered emotion, and the triggered special effect.

[0227] Further, the dynamic special effect processing device further includes:

[0228] A response module, configured to respond to user role preferences;

[0229] A fifth determination module, configured to determine the trigger timing and the triggered emotional special effects according to the user role preferences, the plot turning points, the emotional trajectory of the character, and the barrage analysis result corresponding to the video to be processed.

[0230] The beneficial effects of the embodiments of the present application: in-depth plot analysis, emotion reasoning, and emotional trajectory construction of the video to be processed, to accurately analyze the video plot and emotional fluctuations, and obtain the emotional intensity, the emotion-related character, the plot turning points, and the barrage analysis result corresponding to the video to be processed, determine the trigger timing, the triggered emotion, and the triggered special effect, perform emotional special effect rendering through the trigger timing, the triggered emotion, and the triggered special effect to achieve natural rendering, and perform emotional special effect rendering by controlling the rendering duration and intensity of the special effect according to the emotional intensity curve to ensure that the special effect display is consistent with the plot development. At the same time, perform emotional special effect rendering through the trigger timing, the triggered emotion, and the triggered special effect. When multiple emotions coexist, it can smoothly transition, alternate multiple emotional special effects, and connect naturally, avoiding visual conflicts, and achieving accurate triggering and rendering efficiency of the emotional special effects. In addition, provide a personalized special effect experience according to the user's emotional preferences, effectively enhancing the audience's immersion and improving the user's viewing experience.

[0231] The embodiments of the present application further provide a storage medium, where the storage medium includes stored instructions, and when the instructions run, the device where the storage medium is located is controlled to execute the dynamic special effect processing method as described above.

[0232] The embodiments of the present application further provide an electronic device, the structural schematic diagram of which is as Figure 4 shown, specifically including a memory 401, and one or more instructions 402, where one or more instructions 402 are stored in the memory 401 and are configured to be executed by one or more processors 403 to execute the above-mentioned dynamic special effect processing method.

[0233] For the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0234] It should be noted that the various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other. For system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the relevant parts.

[0235] The steps in the methods of the various embodiments of this application can be adjusted, combined, and deleted according to actual needs.

[0236] Finally, it should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0237] The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the scope of this application. Therefore, this application will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0238] The foregoing is only the preferred embodiment of this application. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this application.

Claims

1. A dynamic special effects processing method, characterized in that: The method comprises: Perform plot analysis on the acquired video to be processed to obtain the video plot background, emotional tone and key plot nodes; According to the video plot background and the emotional tone in the plot, emotional reasoning and emotional trajectory construction are performed respectively to obtain emotional labels and emotional trajectories of the characters; Identify plot turning points through the emotional trajectory of the characters and the key plot nodes; Determine the triggering time, triggering emotion and triggering special effect according to the emotion label, the acquired emotion intensity, emotion-related role, the plot turning point and the bullet screen analysis result corresponding to the video to be processed; Emotional special effects rendering is performed through the triggering timing, the triggering emotion and the triggering special effects.

2. The method according to claim 1, characterized in that The process of obtaining key plot nodes includes: The video to be processed is subjected to script line analysis, visual retrieval recognition, character dynamic behavior analysis and emotional tone analysis respectively to complete the process of obtaining key plot nodes.

3. The method according to claim 1, characterized in that According to the video plot background and the emotional tone in the plot, emotional reasoning and emotional trajectory construction are performed respectively to obtain emotional labels and emotional trajectory of the characters, including: Obtaining emotional changes of the character in the video to be processed at various time points; Extracting emotion labels according to the plot background of the video, the emotional tone in the plot, and the emotional changes of the characters at various time points; Based on the emotional tone in the plot, the emotional fluctuations between time points are determined, and based on the emotional fluctuations between the time points, the emotional trajectory of the character is inferred.

4. The method according to claim 3, characterized in that Determining the emotional fluctuations between time points based on the emotional tone in the plot, and inferring the emotional trajectory of the character based on the emotional fluctuations between the time points, includes: For each frame of the video to be processed, extracting emotion intensity data of the same character from the emotion report corresponding to each frame of the video; the emotion intensity data represents the emotion fluctuation of the same character at different time points; Generate a smooth curve of emotion data by using a preset interpolation algorithm and the emotion intensity data of the same character; Establishing an emotion trajectory curve according to the emotion data smoothing curve and the pre-acquired time point data of the video to be processed; Identify emotional fluctuations in the emotional trajectory curve to obtain preset fluctuation moments and emotional fluctuation data; The emotional trajectory of the character is inferred based on the preset fluctuation moments and the emotional fluctuation data.

5. The method according to claim 1, characterized in that The step of identifying plot turning points through the character's emotional trajectory and the key plot nodes includes: According to the emotional trajectory of the character and the key plot nodes, emotion intensity recognition is performed to obtain a recognition result; If the recognition result indicates that the increase value of the emotion intensity within the preset time interval is greater than or equal to the preset value, the time interval corresponding to the recognition result is determined as a plot turning point.

6. The method according to claim 1, characterized in that The rendering of emotional special effects according to the triggering timing, the triggering emotion and the triggering special effects includes: According to the triggering timing, determining the triggering moment of the emotional special effect rendering and the termination moment of the emotional special effect rendering; Calculating the time difference between the triggering moment and the termination moment; Determine the time difference as rendering duration; Emotional special effects rendering is performed according to the rendering duration, the triggering emotion and the triggering special effects.

7. The method according to claim 1, characterized in that Also includes: Respond to user role preferences; The triggering timing and the triggered emotional effects are determined according to the user role preference, the plot turning point, the emotional trajectory of the role and the barrage analysis result corresponding to the video to be processed.

8. A dynamic special effects processing device, characterized in that: The device comprises: An analysis unit is used to perform plot analysis on the acquired video to be processed, and obtain the plot background of the video, the emotional tone of the plot, and key plot nodes; An inference construction unit, used to perform emotion inference and emotion trajectory construction according to the video plot background and the emotional tone in the plot, to obtain an emotion label and an emotion trajectory of the character; An identification unit, used to identify plot turning points through the emotional trajectory of the character and the key plot nodes; A determination unit, configured to determine a triggering time, a triggering emotion, and a triggering special effect according to the emotion label, the acquired emotion intensity, the emotion-related role, the plot turning point, and the bullet screen analysis result corresponding to the video to be processed; The rendering unit is used to perform emotional special effect rendering according to the triggering timing, the triggering emotion and the triggering special effect.

9. A storage medium, characterized in that: The storage medium includes stored instructions, wherein when the instructions are executed, the device where the storage medium is located is controlled to execute the dynamic special effects processing method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It comprises a memory and one or more instructions, wherein the one or more instructions are stored in the memory and are configured to be executed by one or more processors to execute the dynamic special effect processing method as claimed in any one of claims 1 to 7.

Images