Animation generation method and device, equipment and medium

Abstract

The embodiment of the invention discloses an animation generation method and device, equipment and a medium. The scheme can comprise the steps of obtaining text description information for product introduction; performing statement segmentation on the text description information by utilizing a statement segmentation model to obtain a plurality of pieces of statement information; selecting animation segments matched with the statement information from a preset animation segment library to obtain a plurality of animation segments; and splicing the plurality of animation segments according to the sequence of the plurality of pieces of statement information in the text description information to generate an animation corresponding to the text description information.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an animation generation method, device, equipment and medium. Background Art

[0002] Animation is a visual art form that creates moving images by continuously displaying a series of images. It can attract consumers' interest in products in a short period of time by quickly and intuitively conveying the characteristics and advantages of products. It can also introduce products to consumers in a simple and clear manner and improve consumers' understanding of products.

[0003] Efficient animation production is crucial for the rapid promotion of products. Therefore, how to improve the efficiency of animation generation is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The embodiments of this specification provide an animation generation method, apparatus, device and medium to improve the efficiency of animation generation.

[0005] To solve the above technical problems, the embodiments of this specification are implemented as follows:

[0006] An animation generation method provided in an embodiment of the present specification includes: obtaining text description information for product introduction; performing sentence segmentation on the text description information using a sentence segmentation model to obtain a plurality of sentence information; selecting animation clips matching each of the sentence information from a preset animation clip library to obtain a plurality of animation clips; splicing the plurality of animation clips in the order of the plurality of sentence information in the text description information to generate an animation corresponding to the text description information.

[0007] An animation generation device provided in an embodiment of the present specification includes: an information acquisition module, which is used to acquire text description information for product introduction; a sentence segmentation module, which is used to perform sentence segmentation on the text description information using a sentence segmentation model to obtain a plurality of sentence information; an animation clip selection module, which is used to select animation clips matching each of the sentence information from a preset animation clip library to obtain a plurality of animation clips; and an animation clip splicing module, which is used to splice the plurality of animation clips in the order of the plurality of sentence information in the text description information to generate an animation corresponding to the text description information.

[0008] An animation generation device provided in an embodiment of the present specification includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can implement the above-mentioned animation generation method.

[0009] An embodiment of the present specification provides a computer-readable medium on which computer-readable instructions are stored. The computer-readable instructions can be executed by a processor to implement the above-mentioned animation generation method.

[0010] At least one embodiment provided in this specification can achieve the following beneficial effects: the embodiment of this specification can obtain text description information for product introduction; use the sentence segmentation model to segment the text description information into sentences to obtain several sentence information; select animation clips matching each sentence information from the preset animation clip library to obtain several animation clips; splice the several animation clips according to the order of the several sentence information in the text description information to generate an animation corresponding to the text description information. The embodiment of this specification can improve the efficiency of animation generation by selecting animation clips for generating animation from the preset animation clip library and splicing the various animation clips to generate animation.

[0011] On the other hand, the animation clips selected in the embodiment of this specification are matched with the various sentence information of the text description information. The animation clips can be spliced ​​according to the order of the sentence information in the text description information, thereby generating an animation that matches the various sentence information of the text description information. The generated animation has a high degree of matching with the text description information, thereby being able to generate high-quality animation and generate animation that is more suitable for the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0013] Figure 1 This is a schematic diagram of an application scenario of an animation generation method provided by an embodiment of the present application;

[0014] Figure 2 It is a flowchart of an animation generation method provided by an embodiment of the present application;

[0015] Figure 3 is a schematic diagram of an animated background image provided by an embodiment of the present application;

[0016] Figure 4 It is a flowchart of an animation generation method provided by an embodiment of the present application;

[0017] Figure 5 is a schematic diagram of an interface for displaying text description information provided by an embodiment of the present application;

[0018] Figure 6 It is a schematic diagram of an interface for displaying segmented sentence information provided by an embodiment of the present application;

[0019] Figure 7 is a schematic diagram of a text editing area provided by an embodiment of the present application;

[0020] Figure 8 is a schematic diagram of an animation clip matching interface provided by an embodiment of the present application;

[0021] Fig. 9 is a schematic diagram of a text editing area provided by an embodiment of the present application;

[0022] Fig.10 The embodiments of this specification provide corresponding to Figure 2 A structural schematic diagram of an animation generating device;

[0023] Fig.11 The embodiments of this specification provide corresponding to Figure 2 A structural schematic diagram of an animation generating device. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of one or more embodiments of this specification.

[0025] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0026] In the traditional animation production process, it is usually necessary to manually search for relevant animation materials or manually draw animation materials, such as images, animation clips, etc., and then manually process these materials such as editing and splicing to generate animation.

[0027] However, in the above animation production process, searching for image materials usually takes a lot of time, which affects the efficiency of animation generation; and manual processing of materials not only relies on the animation processing technology of professionals but also further limits the efficiency of animation generation.

[0028] In order to solve the defects in the prior art, this solution provides the following embodiments:

[0029] Figure 1This is a schematic diagram of an application scenario of an animation generation method provided in an embodiment of this specification. Figure 1 As shown, the application scenario includes a terminal 1 and a server 2.

[0030] Among them, the terminal 1 can be a smart phone. In a specific embodiment, the terminal 1 can also be one or more of a desktop computer, a laptop computer, a tablet computer, an Internet of Things device, a portable wearable device, or an immersive image display device. Among them, the Internet of Things device can be one or more of a smart speaker, a smart TV, a smart air conditioner, or a smart car device. The portable wearable device can be one or more of a smart watch, a smart bracelet, or a head-mounted device. The immersive image display device includes but is not limited to an augmented reality (AR) device, a virtual reality (VR) device, etc.

[0031] Server 2 can be an independent physical server, or a server cluster or distributed file system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms.

[0032] In the embodiments of the present specification, the terminal 1 and the server 2 can be directly or indirectly connected via wired or wireless communication. The terminal 1 can receive text description information for product introduction input by the user and send it to the server 2. The server 2 has a program that can generate animations based on the text information. The server 2 can use the sentence segmentation model to segment the text description information into sentences to obtain a number of sentence information; select animation clips that match each sentence information from the preset animation clip library to obtain a number of animation clips; splice the several animation clips according to the order of the several sentence information in the text description information to generate an animation corresponding to the text description information. Then the server 2 can send the generated animation to the terminal 1.

[0033] In addition, the terminal 1 itself can also generate animations. For example, the terminal 1 has a program for generating animations. After receiving the text description information for product introduction input by the user, the terminal 1 does not send the text description information to the server 2, but uses the sentence segmentation model to segment the text description information to obtain a number of sentence information; selects animation clips matching each sentence information from the preset animation clip library to obtain a number of animation clips; splices the several animation clips according to the order of the several sentence information in the text description information to generate animations corresponding to the text description information.

[0034] The following describes the animation generation method provided by the embodiments of this specification in conjunction with the accompanying drawings.

[0035] Figure 2 This is a flowchart of an animation generation method provided by an embodiment of the present application. From a program perspective, the execution subject of the process can be a program installed in a server or an animation production platform or terminal. From a hardware perspective, the execution subject of the process can be a server or an animation production platform or terminal that can generate animations. Figure 2 As shown, the method may include the following steps.

[0036] Step 202: Obtain text description information for product introduction.

[0037] In the embodiment of the present specification, the text description information used for product introduction may include at least one of product introduction information and advertising copy information.

[0038] Optionally, the product introduction information may be information used to introduce the product. For example, the product introduction information may include at least one of product name information, product brand information, product technical parameter information, product usage information, product function introduction information, and product usage instructions information.

[0039] Optionally, the advertisement copy information may be information for attracting the user's attention and prompting the user to take specific actions on the product, wherein the specific action may be trial, registration, purchase, etc. of the product.

[0040] As a specific implementation method, the advertising copy information may include at least one of product price information, product discount information, product historical use case information, user historical use feedback information on the product, product target group information, product after-sales service information, and product purchase channel information.

[0041] In the embodiments of this specification, the products corresponding to the text description information may be consumer products such as paper towels, snacks, cosmetics, skin care products, etc., clothing accessories such as clothing, watches, jewelry, etc., electronic products such as mobile phones, headphones, computers, speakers, etc., service products such as online courses, academic tutoring, legal consulting, etc., financial products such as insurance products, financial products, transportation products such as cars, motorcycles, bicycles, etc. Optionally, the products corresponding to the text description information may also be applications, mini-programs, public accounts, etc.

[0042] In the embodiments of this specification, the text description information can be used to generate an animation, which can be an animation for product introduction. In practical applications, users who need to generate animations, such as advertisers, marketing teams, animation producers, and animation publishers, can input text description information for product introduction through a terminal, so that the execution subject of the process can obtain the text description information from the terminal.

[0043] Step 204: Use the sentence segmentation model to perform sentence segmentation on the text description information to obtain a plurality of sentence information.

[0044] Sentence Tokenization is a basic task in Natural Language Processing (NLP). Based on sentence tokenization, a continuous text data can be segmented into individual sentences.

[0045] In the embodiments of this specification, the text description information can be segmented into sentences to obtain a plurality of independent sentence information. Independent sentence information can refer to sentence information that can express a complete idea. In practical applications, the sentence segmentation model for segmenting the text description information can be at least one of a large model, a traditional model, and a sentence segmentation engine.

[0046] As a specific implementation, the big model may refer to a deep learning model that is pre-trained on a large corpus using an autoregressive method in the field of natural language processing. Specifically, the big model may be a model of the GPT series, such as GPT-3.5, GPT-4, and GPT-4o. The big language model may also be a model of other series, such as the Tongyi Qianwen model, the Ant Bailing big model, the Transformer-based model, etc., which are not limited here.

[0047] As a specific implementation, the traditional model may refer to a model other than a large model. The traditional model may be trained based on historical text description information and historical sentence information. Specifically, the traditional model may be at least one of a random forest model, XGBoost (eXtreme Gradient Boosting), a linear regression model, a logistic regression model, a decision tree, a support vector machine (SVM), and an autoregressive integrated moving average model (ARIMA).

[0048] As a specific implementation, the sentence segmentation engine can be a rule set constructed based on a series of rules. These rules represent the conditions for sentence segmentation of text description information. Specifically, the conditions can be at least one of whether the text description information contains preset punctuation marks, whether the text description information contains preset keywords, whether the number of characters contained in the text description information reaches a preset character number threshold, etc. The preset punctuation marks can be commas, periods, semicolons, exclamation marks, colons, and dashes, etc. The preset keywords can be text conjunctions such as "because", "so", "however", "but", modal particles such as "oh", "what", "ah", "maybe", etc.

[0049] In practical applications, the text description information is segmented into sentences, and the obtained multiple sentence information may be one sentence information. The one sentence information may refer to the text description information. Optionally, the text description information is segmented into sentences, and the obtained multiple sentence information may also be two or more sentence information. As a specific implementation method, the text description information may be segmented into sentences manually instead of using a sentence segmentation model.

[0050] Step 206: Selecting animation clips matching each of the sentence information from a preset animation clip library to obtain a plurality of animation clips.

[0051] In practical applications, the preset animation clip library may be set before executing the solution of the embodiment of this specification. In the embodiment of this specification, the animation clip in the preset animation clip library may be a video clip composed of multiple image frames. Specifically, the animation clip may be an animation clip with a preset playback duration, such as 1 second, 2 seconds, 5 seconds, etc. The playback durations of different animation clips may be the same or different. Optionally, the animation clip in the preset animation clip library may also be an image containing an image frame. As a specific implementation, the animation clip may also include audio corresponding to the text description information.

[0052] The embodiments of this specification can generate an animation without audio corresponding to text description information based on a plurality of animation clips, or can generate an animation with audio corresponding to text description information based on a plurality of animation clips.

[0053] Optionally, the animation clips in the preset animation clip library can be obtained from various data sources. For example, they can be downloaded and saved from the Internet. For example, you can enter "animation clip" in the search area of ​​the web page to search for animation clips, or enter "clip description information" in the search area to search for animation clips, and then save the searched animation clips to the preset animation clip library. Specifically, "clip description information" refers to the description information of the animation clip that the user wants to obtain. For example, a clip description information can be "the operation animation of the user placing an order for a product on a mobile phone." The animation clip can also be captured from the existing introduction videos of various products, etc., and the source of the animation clip is not specifically limited here.

[0054] Optionally, the animation clips in the preset animation clip library can be designed by animation designers according to product requirements. For example, advertisers, marketing teams or animation designers can first design clip description information for the desired animation clips, such as the aforementioned "user operation animation of placing an order on a mobile phone", and then the animation designer can design the animation clip based on the clip description information.

[0055] Optionally, the animation clips in the preset animation clip library can also be generated based on the big model. For example, based on the clip description information mentioned above, a prompt word for inputting the big model is constructed, and the prompt word can include the clip description information and reference samples, etc., and then the big model generates the animation clip based on the prompt word.

[0056] In the embodiment of this specification, the animation clips in the preset animation clip library can be animation clips related to each product. In practical applications, the preset animation clip library can include one animation clip or multiple animation clips.

[0057] In the embodiments of this specification, any sentence information can be matched with an animation clip in a preset animation clip library to obtain an animation clip that matches any sentence information. As a specific implementation, the text vectorized data of the sentence information can be matched with the image vectorized data of the animation clip to determine the animation clip that matches the sentence information. As a specific implementation, the sentence information can also be matched with the segment description information of the animation clip to determine the animation clip that matches the sentence information.

[0058] In the embodiment of the present specification, the obtained animation clip matching any of the sentence information may be one animation clip or multiple animation clips. If there are multiple animation clips, the user can select the most suitable one from the multiple animation clips.

[0059] In practical applications, for any sentence information, one animation clip can also be determined from multiple animation clips matching any sentence information based on the context sentence information of any sentence information. For example, one animation clip can be determined based on a first animation clip matching the previous sentence information of any sentence information, and / or a second animation clip matching the next sentence information of any sentence information. Specifically, one animation clip can be determined based on the design style similarity between animation clips. Among them, one animation clip with the highest design style similarity to the first animation clip can be selected from multiple animation clips matching any sentence information as one animation clip. Alternatively, one animation clip with the highest design style similarity to the second animation clip can be selected from multiple animation clips matching any sentence information as one animation clip. Alternatively, one animation clip with the highest design style similarity to the first animation clip and the second animation clip can be selected from multiple animation clips matching any sentence information as one animation clip, and so on.

[0060] Step 208: splicing the plurality of animation clips according to the order of the plurality of sentence information in the text description information to generate an animation corresponding to the text description information.

[0061] In practical applications, the animation clips can be manually spliced ​​into the animation corresponding to the text description information. Optionally, the animation clips can also be spliced ​​into the animation corresponding to the text description information using animation splicing software.

[0062] In the embodiment of this specification, several animation clips are spliced ​​according to the order of several sentence information in the text description information, indicating that the order of a certain animation clip in the spliced ​​animation is consistent with the order of the sentence information corresponding to the certain animation clip in the text description information.

[0063] In practical applications, the generated animation can be used to display to the target audience users of the product. Specifically, the animation can be displayed to the target audience users through online channels, such as search engines, social media, video platforms, various applications, etc. The animation can also be displayed to the target audience users through offline channels, such as television, subway station screens, elevator screens, shopping mall screens, etc.

[0064] The embodiments of this specification can select several animation clips for generating animation from a preset animation clip library, saving time for manual design and searching of animation clips, improving the efficiency of obtaining animation clips, and splicing several animation clips into an animation, thereby improving the efficiency of animation generation.

[0065] On the other hand, the animation clips selected in the embodiments of this specification are matched with the sentence information after the text description information is segmented into sentences. By splicing several animation clips according to the order of several sentence information in the text description information, an animation matching the various sentence information of the text description information can be generated, so that the generated animation has a high degree of matching with the text description information, thereby being able to generate high-quality animation.

[0066] Furthermore, the solution of the embodiment of this specification does not require the user to have animation production technology, for example, the user may not have the ability to write animation code or the ability to draw animation. The user inputs the text description information for product introduction, and then the animation corresponding to the text description information can be obtained based on the solution of the embodiment of this specification, which makes animation generation more convenient and thus facilitates improving the efficiency of animation generation.

[0067] In the related art, Vincent video models such as Sora and Kling can generate animation content based on text information input by users. However, since Vincent video models cannot semantically identify the detailed content of the industry in which the product is located, the text video models generally generate animations of common styles, such as two-dimensional style, ancient style, literary style and other common styles, but cannot generate animations closely related to the product, resulting in poor quality of the generated animation. In the embodiments of this specification, according to the text description information input by the user, animation clips related to the product that match the text description information can be selected from the preset animation clip library, so that the animation clips can be spliced ​​into animations closely related to the product, which can improve the quality of animation generation.

[0068] It should be understood that the order of some steps in the methods described in one or more embodiments of this specification can be interchanged according to actual needs, or some steps can be omitted or deleted.

[0069] based on Figure 2 The method, the examples of this specification also provide some specific implementation plans of the method, which are described below.

[0070] In the embodiments of this specification, in order to obtain an animation with better effects and more details, the number of characters represented by the text description information can be greater than or equal to a preset number threshold. Specifically, the user can input text description information greater than or equal to the preset number threshold, so that the text description information can be divided into more sentence information, and then more animation clips can be matched, so that the animation clips can be spliced ​​into an animation with better effects and more details. Optionally, the preset number threshold can be 20, 50, 80, etc.

[0071] In actual applications, the user may input text description information with a relatively small number of characters. Then, the animation may be generated using the text description information with a relatively small number of characters. Optionally, the number of texts represented by the text description information may be less than a preset number threshold.

[0072] As a specific implementation, in order to obtain an animation with better effects and more details, the embodiment of this specification may also expand the text description information. Optionally, the method may also include: based on the semantic information or text structure information of the text description information, expanding the text description information to obtain an expanded text. The number of characters contained in the expanded text is greater than the number of characters contained in the text description information.

[0073] The using the sentence segmentation model to perform sentence segmentation on the text description information may specifically include: using the sentence segmentation model to perform sentence segmentation on the expanded text.

[0074] In the embodiments of this specification, the semantic information of the text description information may refer to the meaning or intention expressed by the words and sentences in the text description information. Specifically, the meaning or intention expressed by the words and sentences can be obtained through the literal meaning of the words and sentences, the emotions and tones expressed by the words and sentences, the context of the words and sentences, and the background knowledge related to the words and sentences.

[0075] In the embodiments of the present specification, the text structure information of the text description information may refer to the arrangement order of words and sentences in the text description information. The text structure information of the text description information may represent the presentation method or understanding method of the text description information.

[0076] Optionally, the text structure information of the text description information may refer to the syntactic structure of the text description information. Specifically, the syntactic structure may be a syntactic component, such as a subject, a predicate, an object, an attributive, an adverbial modifier, etc. The syntactic structure may also be a sentence type, such as a simple sentence, a parallel sentence, a compound sentence, etc. For example, a simple sentence may be a sentence containing only one subject and one predicate, such as "I am studying"; a parallel sentence may be a sentence connected by more than one simple sentence, such as "I am studying, he is playing"; a compound sentence may be a sentence containing a subject and at least one subordinate clause, wherein the subordinate clause indicates a sentence that cannot express the complete meaning independently, but serves as a part of the main sentence and serves to decorate or supplement the main sentence. A compound sentence may be "I am studying because I have an exam", "I am studying" is the main clause, and "because I have an exam" is the subordinate clause.

[0077] In practical applications, large models can be used to expand text description information. Since large models are trained on the basis of large-scale corpus data sets, they have good natural language processing capabilities. Based on the semantic information or text structure information of text description information, large models can expand text description information to obtain high-quality expanded text.

[0078] In order to facilitate the understanding of the text description information and the expanded text, the embodiments of this specification also provide specific examples of the text description information and the expanded text. For example, the text description information may be, "Smart home system A can control various smart home devices based on voice recognition technology."

[0079] The expanded text can be, "With the rapid development of science and technology, smart home has become a standard feature of modern families. In order to allow more people to enjoy the convenience brought by science and technology, we have launched a new generation of smart home system - A. The A product has built-in advanced voice recognition technology, and you can control home appliances with just one sentence. Since installing A, my life has become more convenient and comfortable - this is a real evaluation from users. The future is here, and smart life is within reach. Choose A and let technology bring more beauty to your life!".

[0080] For example, the text description information may be, "An accident occurred on the way to work." The expanded text may be, "A pedestrian had an accident with a vehicle on the way to work, resulting in the pedestrian lying on the ground with an injured knee and a bicycle overturned on the road."

[0081] In the embodiment of this specification, the number of characters included in the expanded text may be greater than the number of characters included in the text description information. Therefore, by performing sentence segmentation on the expanded text, more sentence information can be obtained, so that more animation clips can be matched, and animations with better effects and more details can be obtained based on the animation clips.

[0082] In practical applications, punctuation marks can play the role of pause, separation, ending, etc. in the text. Therefore, the text description information can be segmented into sentences based on the punctuation marks in the text description information. Optionally, the sentence segmentation of the text description information using the sentence segmentation model can specifically include: determining the punctuation marks contained in the text description information.

[0083] The text description information is segmented based on the punctuation marks.

[0084] In the embodiment of the present specification, the punctuation marks included in the text description information may be at least one of a comma, a colon, a period, a semicolon, an exclamation mark, a colon, a quotation mark, an ellipsis, a bracket, and a dash.

[0085] As a specific implementation, punctuation recognition tools such as optical character recognition (OCR) software may be used to recognize punctuation in text description information.

[0086] In the embodiments of this specification, the sentence segmentation model can be used to segment the text description information based on punctuation marks. Specifically, for example, for a certain punctuation mark, the sentence segmentation model can be used to segment the sentence before the punctuation mark, and the sentence after the punctuation mark, etc.

[0087] In the embodiments of this specification, the sentence segmentation model can be trained based on sentence samples containing sentence information boundaries marked. The sentence segmentation model can be a support vector machine model, a decision tree model, an LSTM model, a Transformer model, etc. The sentence segmentation model can identify sentence information boundaries in sentences. Furthermore, the embodiments of this specification can segment sentences before or after punctuation marks based on sentence information boundaries.

[0088] In practical applications, the words and sentences before and after the semicolon, comma, etc. may express similar meanings. For example, "This product is very practical and easy to use. Hurry up and recommend it to your friends and partners!" In the text description information, the "practical" before the semicolon and the "easy to use" after the semicolon both mean that the product is suitable for users; the "friends" before the comma and the "partners" after the comma both mean the people around. Therefore, in order to accurately segment sentences and match accurate animation clips to generate high-quality animations, in the embodiment of this specification, two adjacent sentence information with high similarity can be merged into one sentence information. Optionally, after the text description information is segmented based on the punctuation mark, the method may also include: judging whether the semantic similarity of two adjacent text particles is greater than or equal to a preset threshold; the text particle is a subtext obtained by segmenting the text description information based on the punctuation mark. If the similarity is greater than or equal to the preset threshold, the two adjacent text particles are merged into one text particle.

[0089] In the embodiments of this specification, text granules may refer to sentence information obtained by sentence segmentation of text description information. Granules may refer to the granularity of sentence information after sentence segmentation of text description information. For example, sentence information before sentence merging is the first granularity, and sentence information after sentence merging is the second granularity, and the second granularity may be greater than the first granularity.

[0090] In the embodiments of this specification, the sub-text obtained by segmenting the text description information may refer to sentence information of different granularities. For example, the text description information is segmented into sentences to obtain sentence information of the first granularity. If the similarity of two adjacent sentence information of the first granularity is greater than or equal to a preset threshold, the two sentence information are merged into a sentence information of the second granularity. If the similarity of the sentence information of the second granularity and a sentence information of a first granularity adjacent to it is greater than or equal to a preset threshold, the sentence information of the second granularity and the sentence information of the first granularity may be merged into a sentence information of the third granularity. The third granularity may be greater than the second granularity.

[0091] In the embodiments of the present specification, similarity algorithms such as cosine similarity algorithm, semantic similarity algorithm, edit distance algorithm and deep learning-based methods can be used to determine the semantic similarity between two adjacent text particles.

[0092] As a specific implementation, a server, an animation production platform or a terminal may provide a sentence information editing function. A user may edit a plurality of sentence information after sentence segmentation of the text description information, so as to obtain sentence information that better meets product requirements or user requirements. Optionally, the method may further include: displaying the plurality of sentence information. Obtaining a first editing operation of the user for any of the plurality of sentence information. Based on the first editing operation, determining the edited sentence information.

[0093] The selecting the animation clips matching the respective sentence information from the preset animation clip library may specifically include: selecting the animation clips matching the edited sentence information from the preset animation clip library.

[0094] In actual applications, after the text description information is segmented into sentences and several sentence information is obtained, the server, animation production platform or terminal can also display the several sentence information obtained by sentence segmentation, so that the user can view the segmented sentences and can also edit the displayed segmented sentences.

[0095] Optionally, the user can perform a first editing operation on any of the plurality of sentence information. Specifically, the first editing operation can be an operation of modifying the sentence information, such as deleting or replacing a certain word in the sentence information. Alternatively, the first editing operation can also be an operation of deleting the sentence information, or an operation of replacing the sentence information with other sentence information, etc.

[0096] In the embodiment of this specification, after the user performs the first editing operation on the sentence information, the execution subject of the process can select the animation clip that matches the edited sentence information from the preset animation clip library. Because the user edits the sentence information, the sentence information is more in line with the product requirements or user requirements, and then the execution subject of the process can match the animation clip that meets the product requirements or user requirements based on the sentence information, thereby improving the quality of the generated animation.

[0097] In practical applications, the animation clip that matches the sentence information can be determined by calculating the first similarity between the text vectorization data of the sentence information and the image vectorization data of the animation clip. Both the text vectorization data and the image vectorization data are numerical data, so by calculating the first similarity between the text vectorization data and the image vectorization data, it is easy to determine the animation clip that matches the sentence information. Optionally, the preset animation clip library contains multiple animation clips. The method may also include: for any sentence information in each of the sentence information, performing text vectorization processing on the any sentence information to obtain the text vectorization data corresponding to the any sentence information.

[0098] The selecting of the animation clips matching the respective sentence information from the preset animation clip library may specifically include: for any sentence information, calculating a first similarity between the text vectorized data and the picture vectorized data corresponding to the plurality of animation clips, and determining the animation clips matching the any sentence information based on the first similarity.

[0099] In practical applications, at least one of the following algorithms, such as Bag of Words Model (BOW), Term Frequency-Inverse Document Frequency (TF-IDF) algorithm, Word Embedding algorithm, Document to Vector (Doc2vec) algorithm, etc., can be used to perform text vectorization processing on any sentence information.

[0100] Optionally, you can also use Embedding technology to perform text vectorization processing on any sentence information. Specifically, use a machine model with Embedding technology to perform text vectorization processing on any sentence information. Among them, the machine model with Embedding technology can be an Ant Lark model, a BERT model (Bidirectional Encoder Representations from Transformers) or a GPT series model, etc. Embedding technology is a technology that maps high-dimensional data such as text data, image data, etc. to a low-dimensional vector space. The core idea is to convert discrete, unstructured data into continuous vector data for computer processing.

[0101] Optionally, for any of the multiple animation clips, the Embedding technology can be used to extract features of the animation clip to obtain the image vectorization data of the animation clip, such as using the machine model with the Embedding technology mentioned above to extract features of the animation clip.

[0102] In practical applications, for any animation clip among the multiple animation clips, a feature extraction algorithm may be used to extract features of the animation clip to obtain a vectorized digitized image of the animation clip.

[0103] In the embodiments of the present specification, the image vectorization data corresponding to the animation clip can be obtained based on the animation frames in the animation clip. Specifically, for any animation clip among the multiple animation clips, some animation frames, such as one animation frame, two animation frames, etc., can be extracted from the any animation clip in advance, or all animation frames can be extracted. Afterwards, the extracted animation frames can be processed by image vectorization to obtain the image vectorization data of the animation frames. Using the image vectorization data of the animation frames, the image vectorization data corresponding to the any animation clip is obtained. Optionally, animation frame extraction tools such as OpenCV, FFmpeg, etc. can be used to extract some or all animation frames. Alternatively, animation frame interception tools or manual methods can be used to intercept some or all animation frames. Optionally, feature extraction algorithms such as Scale-Invariant Feature Transform (SIFT), Histogram of Oriented Gradients (HOG), accelerated version of the Speed ​​Up Robust Features (SURF), etc. can be used to extract features of the animation frames to obtain feature vectors of the animation frames, which can be image vectorized data of the animation frames.

[0104] As a specific implementation, if an animation frame is extracted from any animation clip, the image vectorization data of the animation frame can be used as the image vectorization data corresponding to the any animation clip. As a specific implementation, if multiple animation frames are extracted from any animation clip, the image vectorization data of the multiple animation frames can be spliced, weighted summed, etc. to obtain the image vectorization data corresponding to the any animation clip.

[0105] In practical applications, the image vectorization data corresponding to the animation clip can also be obtained based on the key frame animation in the animation clip. Optionally, the method can also include: for each of the multiple animation clips, determining the key frame animation contained in each animation clip. The key frame animation is subjected to image vectorization processing to obtain the image vectorization data corresponding to each animation clip.

[0106] In the embodiments of this specification, a key frame animation may represent an animation frame containing the main content of an animation clip. Specifically, a key frame, which may also be an I frame (Intra-coded frame), refers to an image frame in a video that can represent the video content or action changes. In the embodiments of this specification, a key frame animation may be a key frame in an animation clip.

[0107] The embodiments of this specification may use a variety of methods to determine the key frame animation contained in the animation clip. As a specific implementation, the animation clip may be sampled at a preset time interval, and the sampled animation frames may be used as the key frame animation. The preset time interval may be 0.1 seconds, 0.2 seconds, 0.5 seconds, etc., and of course may also be a certain time interval set according to the playback duration of the animation clip.

[0108] As a specific implementation, a key frame animation can be determined from an animation clip based on a frame difference method. Specifically, the frame difference method can calculate the difference between a current frame and an adjacent frame, and if the difference is greater than or equal to a preset difference threshold, the current frame can be used as a key frame animation.

[0109] As a specific implementation, key frame animations can also be determined from animation clips based on clustering algorithms. Specifically, the animation frames in the animation clips can be clustered into different clusters using a K-means clustering algorithm or a hierarchical clustering algorithm. The animation frames in the middle of each cluster on the timeline are used as key frame animations. In practical applications, other methods can also be used to determine key frame animations, such as determining key frame animations based on a convolutional neural network model that identifies key frame animations, and the like.

[0110] In the embodiments of this specification, the determined key frame animation may be a frame of image. For example, the frame of image is subjected to image vectorization processing to obtain the image vectorization data of the frame of image, and then the image vectorization data of the frame of image may be used as the image vectorization data of the animation segment corresponding to the frame of image.

[0111] Optionally, the determined key frame animation may be a plurality of frames of images. For example, the plurality of frames of images are subjected to image vectorization processing to obtain a plurality of image vectorization data corresponding to the plurality of frames of images, and one frame of image corresponds to one image vectorization data. Afterwards, the plurality of image vectorization data corresponding to the plurality of frames of images may be fused, such as weighted fusion of the plurality of image vectorization data, or direct splicing of the plurality of image vectorization data, etc., to obtain the image vectorization data of the animation clip corresponding to the plurality of frames of images.

[0112] In the embodiments of the present specification, when performing image vectorization processing on a key frame animation, the key frame animation can be preprocessed first, such as normalization processing, and then feature extraction algorithms such as scale-invariant feature transformation, oriented gradient histogram, accelerated version of feature algorithms with robust characteristics, etc. can be used to extract features of the key frame animation to obtain a feature vector of the key frame animation, and then the feature vector can be used as the image vectorization data of the key frame animation.

[0113] As a specific implementation, the key frames of the animation clip may not be determined, but the image vectorization data corresponding to the animation clip may be determined directly based on some animation frames in the animation clip, or all animation frames of the animation clip. The specific process of determining the image vectorization data corresponding to the animation clip based on some animation frames or all animation frames can be referred to the content of obtaining the image vectorization data based on multiple frames of images, which will not be described in detail here.

[0114] Optionally, after obtaining the image vectorization data corresponding to each animation clip, the embodiment of this specification can also save the image vectorization data to the preset animation clip library. Therefore, when the solution of the embodiment of this specification is subsequently implemented, it is no longer necessary to calculate the image vectorization data corresponding to the animation clip, thereby improving the calculation efficiency of the first similarity between the text vectorization data and the image vectorization data, thereby improving the efficiency of selecting the animation clip.

[0115] In practical applications, the first similarity may be calculated by using methods for calculating similarity, such as cosine similarity, Euclidean distance, Manhattan distance, Jaccard similarity, and Pearson Correlation Coefficient.

[0116] In the embodiments of the present specification, based on the first similarity, the animation clip that matches any sentence information is determined. Specifically, the animation clip with the highest first similarity can be used as the animation clip that matches any sentence information, or the animation clip with the first similarity greater than or equal to a preset similarity threshold can be used as the animation clip that matches any sentence information, etc.

[0117] Optionally, the first relationship between the image vectorization data of the animation clip and the animation clip can also be saved in a preset animation clip library. In the process of determining the animation clip matching any of the sentence information by the first similarity, the image vectorization data matching any of the sentence information can be determined by the first similarity, and then the animation clip corresponding to the image vectorization data can be determined by the first relationship, thereby determining the animation clip matching any of the sentence information.

[0118] In the embodiment of the present specification, the preset animation clip library may also include clip description information for describing each of the multiple animation clips, so that the animation clip matching the sentence information can be selected based on the clip description information.

[0119] The step of selecting the animation clips matching the sentence information from the preset animation clip library may specifically include:

[0120] For any of the sentence information, a second similarity between the text vectorized data and the segment description vectorized data of each of the segment description information is calculated. Based on the second similarity, a cartoon segment matching the any of the sentence information is determined.

[0121] In the embodiments of this specification, the segment description information may be the description information used to generate the animation segment in the preset animation segment library. For example, the animation segment may be downloaded by the user by inputting the segment description information in a search engine on the Internet; or the animation segment may be obtained by an animation designer through animation design based on the segment description information; or the animation segment description information may be generated by using a large model based on the prompt words containing the segment description information, etc.

[0122] Specifically, the segment description information may be information obtained before the animation segment in the preset animation segment library is generated.

[0123] As a specific implementation, the segment description information may also be obtained after the animation segment in the preset animation segment library is generated. For example, the animation segment may be randomly downloaded from the Internet, and the randomly downloaded animation segment is saved in the preset animation segment library. Then, the segment description information is generated for the randomly downloaded animation segment in the preset animation segment library. Specifically, the segment description information may be generated manually, or by a large model, and so on.

[0124] In practical applications, the second similarity can be calculated by using methods for calculating similarity, such as cosine similarity, Euclidean distance, Manhattan distance, Jaccard similarity, and Pearson correlation coefficient.

[0125] In the embodiments of the present specification, based on the second similarity, the animation clip that matches any sentence information is determined. Specifically, the animation clip with the highest second similarity can be used as the animation clip that matches any sentence information, or the animation clip with a second similarity greater than or equal to a preset similarity threshold can be used as the animation clip that matches any sentence information, etc.

[0126] Optionally, the second relationship between the segment description information of the animation segment and the animation segment can also be saved in a preset animation segment library. In the process of determining the animation segment matching any of the sentence information by the second similarity, the segment description information matching any of the sentence information can be determined by the second similarity, and then the animation segment corresponding to the segment description information can be determined by the second relationship, thereby determining the animation segment matching any of the sentence information.

[0127] In an embodiment of the present specification, the preset animation segment library may also include segment description vectorized data of segment description information of the animation segment. Optionally, the method may also include: performing text vectorization processing on the segment description information of each animation segment to obtain segment description vectorized data of each animation segment. Saving the segment description vectorized data to the preset animation segment library.

[0128] Optionally, at least one algorithm selected from the bag-of-words model, word frequency-inverse document frequency algorithm, word embedding algorithm, document-vector model, etc. may be used to perform text vectorization on the segment description information of each animation segment to obtain segment description vectorized data of each animation segment.

[0129] In an embodiment of the present specification, the fragment description vectorized data can be pre-added to a preset animation fragment library, so that when the scheme of the embodiment of the present specification is subsequently implemented, there is no need to calculate the fragment description vectorized data of the fragment description information of the animation fragment, thereby improving the calculation efficiency of the second similarity between the text vectorized data and the fragment description vectorized data, thereby improving the selection efficiency of the animation fragment.

[0130] Optionally, the preset animation clip library may also include a third relationship between the segment description vectorized data of the segment description information of the animation clip and the animation clip. In the process of determining the animation clip that matches any of the sentence information by the second similarity, the segment description vectorized data that matches any of the sentence information may be determined by the second similarity, and then the animation clip that matches any of the sentence information may be determined by the third relationship.

[0131] In the embodiments of the present specification, in order to accurately determine the animation clips that match any of the sentence information, the first similarity and the second similarity may be combined to determine the animation clips that match any of the sentence information. Optionally, the determination of the animation clips that match any of the sentence information based on the second similarity may specifically include: based on the first similarity, determining a first animation clip set from the multiple animation clips whose first similarity is greater than or equal to a first preset threshold. Based on the second similarity, determining a second animation clip set from the multiple animation clips whose second similarity is greater than or equal to a second preset threshold. Determine the intersection of the first animation clip set and the second animation clip set. Determine the animation clips contained in the intersection as the animation clips that match any of the sentence information.

[0132] Optionally, the first animation clip set may include at least one animation clip. Optionally, the second animation clip set may include at least one animation clip.

[0133] In the embodiment of this specification, the intersection of the first animation clip set and the second animation clip set may include one animation clip. The one animation clip may be used as the animation clip matching any of the sentence information. As a specific implementation, the intersection of the first animation clip set and the second animation clip set may include two or more animation clips. One animation clip may be selected from the two or more animation clips as the animation clip matching any of the sentence information.

[0134] Specifically, if the intersection includes two or more animation clips, one animation clip can be randomly selected from the two or more animation clips as the animation clip that matches any of the sentence information. Alternatively, the first similarity and the second similarity of the two or more animation clips can be weighted to obtain a third similarity, and the animation clip with the highest third similarity can be used as the animation clip that matches any of the sentence information. Alternatively, the first preset threshold and / or the second preset threshold can be increased, the first animation clip set and / or the second animation clip set can be re-determined, and then the intersection can be re-determined, and the animation clip that matches any of the sentence information can be re-determined, and so on.

[0135] As a specific implementation, if the intersection includes two or more animation clips, the two or more animation clips can be spliced ​​with the animation clips that match other sentence information, thereby generating multiple animations. The user can select an animation that meets his or her needs from the multiple animations.

[0136] In the embodiment of this specification, the intersection of the first animation clip set and the second animation clip set may also be an empty set.

[0137] Specifically, if the intersection is an empty set, the animation clip with the highest first similarity can be selected from the first set to be determined as the animation clip matching any of the sentence information; or the animation clip with the highest second similarity can be selected from the second set to be determined as the animation clip matching any of the sentence information. Alternatively, the first preset threshold and / or the second preset threshold can be lowered, the first animation clip set and / or the second animation clip set can be re-determined, and then the intersection can be re-determined, and the animation clip matching any of the sentence information can be re-determined, and so on.

[0138] In the embodiments of this specification, a server, an animation production platform or a terminal may provide an animation clip editing function. Thus, a user may also edit an animation clip selected from a preset animation clip library that matches each sentence information, and may obtain an animation clip that better meets product requirements or user requirements. Optionally, the method may further include: displaying the plurality of animation clips. Obtaining a second editing operation of the user for any of the plurality of animation clips. Based on the second editing operation, determining the edited animation clip.

[0139] The step of splicing the plurality of animation clips according to the order of the plurality of sentence information in the text description information may specifically include: splicing the edited animation clips according to the order.

[0140] In actual applications, the animation clips selected from the preset animation clip library and matching the respective sentence information can be displayed in the animation production interface of the server or animation production platform.

[0141] In the embodiment of the present specification, the user can perform a second editing operation on any animation clip among several animation clips. Specifically, the second editing operation can be an operation of modifying a certain animation frame in the animation clip, such as modifying an image or text in a certain animation frame, etc. The second editing operation can also be an operation of deleting the animation clip, or an operation of replacing the animation clip with another animation clip, etc.

[0142] In the embodiment of the specification, after the second editing operation is performed on the animation clips, the edited animation clips can be spliced ​​according to the order of the sentence information corresponding to the edited animation clips in the text description information to generate an animation. Since the second editing operation is performed on the animation clips, the quality of the animation clips can be improved, and animation clips that meet user needs can be obtained, thereby generating high-quality animations.

[0143] In the embodiments of this specification, in order to improve the quality of the animation, the sentence information can also be synchronized to the animation segment in the animation that matches the sentence information. Optionally, the method can also include: synchronously displaying each of the sentence information in the matching animation segment. The generating of the animation corresponding to the text description information can specifically include: generating an animation containing the text description information.

[0144] In actual applications, the sentence information can be synchronously displayed in the matching animation segment according to the playing time of the animation segment matching the sentence information. As a specific implementation, all the text of the sentence information can be displayed in the matching animation segment at once, or all the text of the sentence information can be displayed word by word in the matching animation segment.

[0145] Optionally, the sentence information may be synchronously displayed in any area of ​​the matching animation clip, for example, the sentence information may be synchronously displayed in the upper area, the middle area, the lower area, etc. of the matching animation clip.

[0146] The embodiment of the present specification synchronously displays each sentence information in a matching animation clip, so that the animation clips can be spliced ​​to generate an animation containing text description information, which can enrich the animation content and improve the animation quality.

[0147] In practical applications, in order to further improve the quality of animation, the text description information may be dubbed. Optionally, the method may further include: dubbing the sentence information in the animation clip to obtain dubbing information of the animation clip.

[0148] The generating of the animation including the text description information may specifically include: generating the animation including the text description information and the dubbing information.

[0149] Optionally, the sentence information in the animation clip may be dubbed manually; or dubbing software may be used to dub the sentence information in the animation clip.

[0150] In practical applications, in order to improve the animation generation effect, the animation can also be placed in an animation background picture. Optionally, the method can also include: obtaining an animation background picture; the animation background picture includes a first area for displaying the animation. The animation is integrated with the animation background picture to obtain an animation including the animation background picture.

[0151] In a specific embodiment, the animation background image can be a static image or a dynamic image. In a specific embodiment, the animation background image can be an image related to several animation clips in the animation or an image unrelated to several animation clips in the animation, which is not limited here.

[0152] Optionally, integrating the animation with the animation background image may be placing the animation in the first area of ​​the animation background image.

[0153] Specifically, the first area in the animation background picture may include the central area in the animation background picture, so that the animation can be placed in the central area in the animation background picture to attract the user's attention.

[0154] In the embodiments of this specification, the animation can be placed in the animation background image through animation production software such as PowerPoint, Adobe AfterEffects, Jianying and other software.

[0155] In practical applications, background text may be set in the background animation image to enrich the animation content. Optionally, the animation background image also includes a second area for displaying background text; the background text includes prompt information related to the product.

[0156] In the embodiment of the present specification, the second area may be any area in the animation background image except the first area.

[0157] The background copy may include prompt information related to the product. Specifically, the prompt information may be at least part of the information in the text description information. For example, the prompt information may be product name information, product brand information, a sentence in the text description information, etc. As a specific implementation method, for financial products such as insurance products, financial products, etc., the prompt information may also be compliance information. Compliance information may refer to the laws, regulations, rules and regulations, and industry standards followed by the product. The background copy may be fixedly displayed in the second area; it may also be scrollingly displayed in the second area, which is not limited here.

[0158] Figure 3 is a schematic diagram of an animated background image provided by an embodiment of the present application. Figure 3 As shown, the animation background image may include a first area 302 and a second area 304. The first area 302 may be used to place the generated animation, and the second area 304 may be used to place prompt information. The placed prompt information may include sentence information in the text description information, product name information, product brand information, product compliance information, and product precautions information. Figure 3 The second area 304 may include a first placement area 306 for placing sentence information in the text description information, a second placement area 308 for placing product name information, a third placement area 310 for placing product brand information, a fourth placement area 312 for placing product compliance information, and a fifth placement area 314 for placing precautions information. Figure 3 The first region 302 in the figure is horizontal. In a specific embodiment, the first region 302 may also be vertical.

[0159] As a specific implementation, the animation background image may include one area for placing the animation, or may include multiple areas for placing the animation, which is not limited here.

[0160] In the embodiment of this specification, the text description information may be at least one of the introduction information and the advertising copy information for the product. Specifically, the introduction information of the product may include at least one of the product name information, the brand information of the product, the technical parameter information of the product, the use information of the product, the function introduction information of the product and the instruction information of the product.

[0161] Specifically, the advertising copy information may include at least one of product price information, product discount information, product historical use case information, user historical use feedback information on the product, product target group information, product after-sales service information, and product purchase channel information.

[0162] In the embodiments of this specification, the text description information can be information about various products, such as consumer goods such as paper towels, snacks, etc., clothing accessories such as clothing, jewelry, etc., electronic products such as mobile phones, computers, etc., service products such as academic tutoring, legal consulting, etc., financial products such as insurance products, financial products, etc.

[0163] Figure 4 FIG. 1 is a flow chart of an animation generation method provided by an embodiment of the present application. Figure 4 As shown, the method may include the following steps.

[0164] Step 402: Obtain text description information.

[0165] In the embodiment of this specification, the user can input text description information for product introduction through the terminal. The server as the execution subject of the process can obtain the text description information for product introduction from the terminal.

[0166] Figure 5 is a schematic diagram of an interface for displaying text description information provided by an embodiment of the present application. Figure 5As shown, the interface 500 for displaying text description information may include a display area 502 for displaying text description information. The user may input text description information in the display area 502, so that the server obtains the text description information. Alternatively, the server may also obtain the text description information from other places where the text description information is stored, such as obtaining the text description information from other interfaces or from a database of a text provider, so as to display the text description information in the display area 502. Optionally, the text description information in the embodiment of this specification may include at least one of introduction information and advertising copy information for insurance products.

[0167] Step 404: Segment the sentences.

[0168] In practical applications, the text description information can be sentence segmented using a sentence segmentation model or manually to obtain one or more sentence information. In the embodiment of this specification, the interface 500 for displaying the text description information may also include a sentence segmentation control for sentence segmentation of the text description information. The user can sentence segment the text description information based on the sentence segmentation control. For example, the sentence segmentation control may include a first control 504. The name of the first control 504 may be "AI formatting". The user may click on the first control 504 so that the server may sentence segment the text description information using the sentence segmentation model. The sentence segmentation control may also include a second control 506. The name of the second control 506 may be "manual formatting". The user may click on the second control 506 so that the text description information may be manually sentence segmented.

[0169] In actual applications, the interface 500 for displaying text description information may include a sentence segmentation control, such as the first control 504 or the second control 506 , or may include two sentence segmentation controls, such as the first control 504 and the second control 506 .

[0170] In the embodiment of the present specification, the interface 500 for displaying text description information may further include an operation confirmation control 508. The name of the operation confirmation control 508 may be "Generate Data". The operation confirmation control 508 is used for the user to confirm the selection operation for the sentence segmentation control. After the user selects the sentence segmentation control, the user may click the operation confirmation control 508, so as to obtain the sentence information after the sentence segmentation is performed using the sentence segmentation method corresponding to the sentence segmentation control selected by the user.

[0171] Optionally, in order to improve the quality of the sentence information, the server may also edit the sentence information based on the user's first editing operation on the sentence information.

[0172] Figure 6 is a schematic diagram of an interface for displaying segmented sentence information provided by an embodiment of the present application. Figure 6As shown, the interface 600 for displaying segmented sentence information may include each sentence information 602 after the text description information is segmented. The interface 600 for displaying segmented sentence information may also include an editing control 604. One sentence information 602 may correspond to one editing control 604. The user may click on the editing control 604 to perform a first editing operation on the sentence information 602 corresponding to the editing control 604.

[0173] Assuming that the user wants to edit the sentence information "ABCDEF is a high-quality product launched by abc, a big brand, and guaranteed.", the user can click the first editing control 604 corresponding to the sentence information. In response to the user's click operation, the server can display a text editing area for editing the sentence information in the interface 600 displaying the segmented sentence information.

[0174] Figure 7 Schematic diagram of a text editing area provided by an embodiment of the present application. Figure 7 As shown, the text editing area 702 may include sentence information to be edited. The user may perform a first editing operation on the sentence information in the text editing area 702, such as adjusting the size of the text in the sentence information, adjusting the color of the text in the sentence information, modifying the text content in the sentence information, and the like.

[0175] Step 406: Extract text vectorized data.

[0176] In the embodiment of this specification, the server can perform vectorization processing on each sentence information, thereby extracting text vectorization data of the sentence information. Optionally, the text vectorization data can be used to match the animation clip.

[0177] Step 408: Matching animation clips.

[0178] In the embodiments of this specification, the animation clips may be animation clips in a preset animation clip library. The preset animation clip library may be set before executing the solution of the embodiments of this specification. Optionally, the animation clips in the preset animation clip library may be deposited animation clips. For example, they may be downloaded and saved from the Internet, captured from existing introduction videos of various products, designed by animation designers according to product requirements, or generated based on large models, etc.

[0179] In practical applications, the preset animation clip library may also include at least one of the following: image vectorized data corresponding to each animation clip, segment description information corresponding to each animation clip, and segment description vectorized data of the segment description information corresponding to each animation clip. Optionally, the server may match the text vectorized data with at least one of the image vectorized data, segment description information, and segment description vectorized data in the preset animation clip library to obtain animation clips with matching sentence information.

[0180] Figure 8 is a schematic diagram of an animation clip matching interface provided by an embodiment of the present application. Figure 8 As shown, after matching the animation clips, an animation clip matching interface 800 may also be displayed. The animation clip matching interface 800 may include a sentence information and the animation clip corresponding to the sentence information. The animation clip matching interface 800 may also include multiple sentence information or all sentence information.

[0181] Optionally, the animation clip matching interface 800 may automatically play the animation clip corresponding to the sentence information, and perform special display, such as highlighting, on the sentence information corresponding to the played animation clip.

[0182] Optionally, the user may click on the sentence information in the animation segment matching interface 800, and the animation segment matching interface 800 may perform special display on the sentence information clicked by the user, such as highlighting. The animation segment matching interface 800 may also play the animation segment corresponding to the sentence information clicked by the user. For example, when the user clicks on the sentence information 802, the animation segment matching interface 800 highlights the sentence information 802, and the animation segment matching interface 800 may also play the animation segment 804 corresponding to the sentence information 802.

[0183] Optionally, in order to improve the quality of the animation clip, the server may also edit the matched animation clip based on the second editing operation of the user in the sentence information.

[0184] In the embodiment of this specification, the animation clip matching interface 800 may also include each layer of the animation clip and a picture editing area for each layer. Figure 8 As shown, the animation clip matching interface 800 includes a picture layer 8042 and a text layer 8044 of the animation clip 804. The animation clip matching interface 800 also includes a picture editing area 806 for the layer. The user can perform a second editing operation on the layer based on the picture editing area 806. For example, the user can click on the picture layer 8042, so that the width, height, and position of the picture in the picture layer 8042 can be adjusted in the picture editing area 806.

[0185] In actual applications, the user may also perform a second editing operation on the text in the text layer 8044. For example, the user may click on the text layer 8044, and the animation segment matching interface 800 may display a text editing area.

[0186] Fig. 9 Schematic diagram of a text editing area provided by an embodiment of the present application. Fig. 9 As shown, the user can perform a second editing operation on the text in the text editing area 900, such as adjusting the color and size of the text, modifying the text content, etc. As a specific implementation, the user can also perform a second editing operation in the animation clip. For example, the user can directly click on the text in the animation clip to edit the text in the animation clip.

[0187] Step 410: Perform animation splicing.

[0188] In the embodiment of the present specification, the animation clips can be spliced ​​according to the order of the sentence information in the text description information. Specifically, the animation clips can be spliced ​​manually or by animation splicing software.

[0189] Step 412: Generate audio.

[0190] In the embodiment of the specification, in order to improve the animation quality, the text description information can also be dubbed. Specifically, the text description information can be dubbed after the text description information is segmented into sentences to obtain dubbing information matching the sentence information.

[0191] In the embodiments of the present specification, the sentence information in the animation clip can be dubbed manually; or dubbing software can be used to dub the sentence information in the animation clip.

[0192] Step 414: Generate animation.

[0193] In the embodiment of this specification, the animation obtained in step 410 can be synthesized with the dubbing information obtained in step 412 to obtain the final animation. In practical applications, in order to improve the quality of the animation, the sentence information after the text description information is segmented can also be synchronized to the animation segment matching the sentence information. Thus, an animation containing the text description information is obtained.

[0194] In the embodiments of this specification, in order to enrich the animation content, background text can also be set in the animation. Optionally, the background text can include prompt information related to the product. For example, the prompt information can be a certain sentence information, etc. Optionally, the prompt information can also be the compliance information corresponding to the product. The compliance information can refer to the laws, regulations, rules and regulations, and industry standards followed by the product.

[0195] In the embodiments of this specification, the generated animation can also be saved in a preset animation clip library. The generated animation can be used as an animation clip in the preset animation clip library to facilitate the generation of subsequent animations. As a specific implementation method, the above-mentioned various interfaces may not be displayed. For example, the execution body of the process can provide an interface for obtaining the user's text description information and an interface for displaying animation clips. The user can enter the text description information or paste the text description information in the interface for obtaining the user's text description information, and then the execution body of the process can use the sentence segmentation model in the background to perform sentence segmentation on the text description information, match the sentence information after the sentence segmentation with the animation clip, splice the animation clips, generate the animation corresponding to the text description information, and then the generated animation can be displayed on the interface for displaying the animation clips.

[0196] It should be understood that the order of some steps in the methods described in one or more embodiments of this specification can be interchanged according to actual needs, or some steps can be omitted or deleted.

[0197] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method.

[0198] Fig.10 The embodiments of this specification provide corresponding to Figure 2 A schematic diagram of the structure of an animation generation device. Fig.10 As shown, the apparatus may include:

[0199] The information acquisition module 1002 is used to acquire text description information for product introduction.

[0200] The sentence segmentation module 1004 is used to perform sentence segmentation on the text description information using a sentence segmentation model to obtain a plurality of sentence information.

[0201] The animation clip selection module 1006 is used to select animation clips matching each of the sentence information from a preset animation clip library to obtain a plurality of animation clips.

[0202] The animation segment splicing module 1008 is used to splice the plurality of animation segments according to the order of the plurality of sentence information in the text description information to generate an animation corresponding to the text description information.

[0203] Optionally, the device may further include:

[0204] The text expansion module is used to expand the text description information based on the semantic information or text structure information of the text description information to obtain an expanded text; the number of characters contained in the expanded text is greater than the number of characters contained in the text description information.

[0205] The optional sentence segmentation module 1004 can be specifically used to: perform sentence segmentation on the expanded text using a sentence segmentation model.

[0206] Optionally, the sentence segmentation module 1004 may be specifically configured to: determine punctuation marks included in the text description information, and segment the text description information based on the punctuation marks.

[0207] Optionally, the device may further include: a judgment module, used to judge whether the semantic similarity between two adjacent text particles is greater than or equal to a preset threshold; the text particles are sub-texts obtained by segmenting the text description information based on the punctuation marks.

[0208] The text merging module is used to merge the two adjacent text particles into one text particle if the similarity is greater than or equal to a preset threshold.

[0209] Optionally, the preset animation clip library contains multiple animation clips; the device may also include: a text vectorization processing module, used to perform text vectorization processing on any sentence information in the individual sentence information to obtain text vectorization data corresponding to any sentence information.

[0210] The animation clip selection module 1006 may be specifically configured to: for any of the sentence information, calculate a first similarity between the text vectorized data and the picture vectorized data corresponding to the plurality of animation clips, and determine an animation clip matching the any of the sentence information based on the first similarity.

[0211] Optionally, the device may further include: a key frame animation determination module, configured to determine, for each animation clip in the plurality of animation clips, a key frame animation included in the each animation clip.

[0212] The image vectorization processing module is used to perform image vectorization processing on the key frame animation to obtain the image vectorization data corresponding to each animation segment.

[0213] The image vectorization data saving module is used to save the image vectorization data to the preset animation clip library.

[0214] Optionally, the preset animation clip library further includes clip description information for describing each animation clip in the multiple animation clips.

[0215] The animation segment selection module 1006 may be specifically configured to: for any of the sentence information, calculate a second similarity between the text vectorized data and the segment description vectorized data of each of the segment description information, and determine an animation segment matching the any of the sentence information based on the second similarity.

[0216] Optionally, determining the animation clip matching the any sentence information based on the second similarity may specifically include: determining, from the multiple animation clips, a first set of animation clips whose first similarity is greater than or equal to a first preset threshold, based on the first similarity.

[0217] Based on the second similarity, a second animation clip set whose second similarity is greater than or equal to a second preset threshold is determined from the plurality of animation clips.

[0218] An intersection of the first set of animation clips and the second set of animation clips is determined.

[0219] The animation clips included in the intersection are determined as animation clips matching any of the sentence information.

[0220] Optionally, the device may further include: a segment description vectorized data processing module, configured to perform text vectorization processing on the segment description information of each animation segment to obtain segment description vectorized data of each animation segment;

[0221] The segment description vectorized data saving module is used to save the segment description vectorized data to the preset animation segment library.

[0222] Optionally, the device may further include: a synchronization module, configured to synchronously display each of the sentence information in a matching animation segment.

[0223] The animation segment splicing module 1008 can be specifically used to generate an animation containing the text description information.

[0224] Optionally, the device may further include: an animation background image acquisition module, used to acquire an animation background image; the animation background image includes a first area for displaying the animation.

[0225] The integration module is used to integrate the animation with the animation background image to obtain an animation including the animation background image.

[0226] Optionally, the animation background image also includes a second area for displaying background text; the background text includes prompt information related to the product.

[0227] Optionally, the device may also include: a first display module, used to display the several statement information; a first operation acquisition module, used to obtain a user's first editing operation on any statement information among the several statement information; and an edited statement information determination module, used to determine the edited statement information based on the first editing operation.

[0228] The animation clip selection module 1006 may be specifically configured to select an animation clip that matches the edited sentence information from a preset animation clip library.

[0229] Optionally, the device may also include: a second display module for displaying the several animation clips; a second operation acquisition module for acquiring a second editing operation of a user on any of the several animation clips; and an edited animation clip determination module for determining an edited animation clip based on the second editing operation.

[0230] The animation clip splicing module 1008 may be specifically configured to splice the edited animation clips in the sequence.

[0231] Optionally, the text description information includes at least one of introduction information and advertising copy information for the insurance product.

[0232] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method.

[0233] Fig.11 The embodiments of this specification provide corresponding to Figure 2 A schematic diagram of the structure of an animation generation device. Fig.11 As shown, the device 1100 may include:

[0234] at least one processor 1110; and,

[0235] A memory 1130 is communicatively connected to the at least one processor 1110; wherein,

[0236] The memory 1130 stores instructions 1120 that can be executed by the at least one processor 1110. The instructions 1120 are executed by the at least one processor 1110 so that the at least one processor 1110 can implement the above-mentioned animation generation method.

[0237] Based on the same idea, the embodiment of this specification also provides a computer-readable storage medium corresponding to the above method, on which computer instructions are stored, and when the computer instructions are executed by a processor, the above animation generation method is implemented.

[0238] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. Fig.11 As for the animation generating device shown, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0239] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards in the relevant regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0240] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0241] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.

[0242] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0243] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0244] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0245] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0246] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0247] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0248] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0249] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0250] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0251] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0252] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0253] The present application may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0254] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An animation generation method, comprising: Get text description information for product introduction; Using a sentence segmentation model to segment the text description information into sentences to obtain a plurality of sentence information; Selecting animation clips matching each of the sentence information from a preset animation clip library to obtain a plurality of animation clips; The plurality of animation clips are spliced ​​according to the order of the plurality of sentence information in the text description information to generate an animation corresponding to the text description information.

2. The method of claim 1, further comprising: Expanding the text description information based on the semantic information or text structure information of the text description information to obtain an expanded text; The number of characters included in the expanded text is greater than the number of characters included in the text description information; The sentence segmentation of the text description information using the sentence segmentation model specifically includes: The sentence segmentation model is used to perform sentence segmentation on the expanded text.

3. The method according to claim 1, wherein the sentence segmentation of the text description information using a sentence segmentation model specifically comprises: Determining punctuation marks included in the text description information; The text description information is segmented based on the punctuation marks.

4. The method according to claim 3, after segmenting the text description information based on the punctuation marks, further comprising: Determine whether the semantic similarity between two adjacent text particles is greater than or equal to a preset threshold; The text particles are sub-texts obtained by segmenting the text description information based on the punctuation marks; If the similarity is greater than or equal to a preset threshold, the two adjacent text particles are merged into one text particle.

5. The method according to claim 1, wherein the preset animation clip library comprises a plurality of animation clips; The method further comprises: For any sentence information among the sentence information, perform text vectorization processing on the sentence information to obtain text vectorization data corresponding to the sentence information; The step of selecting the animation clips matching the respective sentence information from the preset animation clip library specifically includes: For any of the sentence information, calculating a first similarity between the text vectorized data and the picture vectorized data corresponding to the plurality of animation clips; Based on the first similarity, a animation clip matching any one of the sentence information is determined.

6. The method of claim 5, further comprising: For each of the plurality of animation clips, determining a key frame animation included in each of the animation clips; Performing image vectorization processing on the key frame animation to obtain the image vectorization data corresponding to each animation segment; The image vectorization data is saved to the preset animation clip library.

7. The method according to claim 5, wherein the preset animation clip library further comprises clip description information for describing each of the plurality of animation clips; The step of selecting the animation clips matching the respective sentence information from the preset animation clip library specifically includes: For any of the sentence information, calculating a second similarity between the text vectorized data and the fragment description vectorized data of each of the fragment description information; Based on the second similarity, a animation clip matching the any sentence information is determined.

8. The method according to claim 7, wherein determining the animation clip matching the any sentence information based on the second similarity specifically comprises: Based on the first similarity, determining, from the plurality of animation clips, a first set of animation clips whose first similarity is greater than or equal to a first preset threshold; Based on the second similarity, determining, from the plurality of animation clips, a second animation clip set whose second similarity is greater than or equal to a second preset threshold; Determine an intersection of the first animation clip set and the second animation clip set; The animation clips included in the intersection are determined as animation clips matching any of the sentence information.

9. The method of claim 7, further comprising: Performing text vectorization processing on the segment description information of each animation segment to obtain segment description vectorized data of each animation segment; The segment description vectorized data is saved to the preset animation segment library.

10. The method of claim 1, further comprising: Synchronously displaying each of the sentence information in a matching animation clip; The generating of the animation corresponding to the text description information specifically includes: Generate an animation containing the text description information.

11. The method of claim 1, further comprising: Get the animated background image; The animation background image includes a first area for displaying the animation; The animation is integrated with the animation background image to obtain an animation including the animation background image.

12. The method according to claim 11, wherein the animation background image further comprises a second area for displaying background text; the background text comprises prompt information related to the product.

13. The method of claim 1, further comprising: Displaying the plurality of statement information; Acquire a first editing operation of a user on any one of the plurality of sentence information; Based on the first editing operation, determining edited sentence information; The step of selecting the animation clips matching the respective sentence information from the preset animation clip library specifically includes: An animation clip matching the edited statement information is selected from a preset animation clip library.

14. The method of claim 1, further comprising: Displaying the plurality of animation clips; Obtaining a second editing operation of the user on any one of the plurality of animation clips; Based on the second editing operation, determining an edited animation clip; The step of splicing the plurality of animation clips according to the order of the plurality of sentence information in the text description information specifically includes: The edited animation clips are spliced ​​in the order.

15. The method according to claim 1, wherein the text description information comprises at least one of introduction information and advertising copy information for the insurance product.

16. An animation generating device, comprising: An information acquisition module is used to acquire text description information for product introduction; A sentence segmentation module, used to perform sentence segmentation on the text description information using a sentence segmentation model to obtain a plurality of sentence information; An animation clip selection module is used to select animation clips matching each of the sentence information from a preset animation clip library to obtain a plurality of animation clips; The animation segment splicing module is used to splice the plurality of animation segments according to the order of the plurality of sentence information in the text description information to generate an animation corresponding to the text description information.

17. An animation generating device, comprising: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can implement the animation generation method according to any one of claims 1 to 15.

18. A computer-readable medium having computer-readable instructions stored thereon, wherein the computer-readable instructions can be executed by a processor to implement the animation generation method according to any one of claims 1 to 15.

Images