A method of generating a video storyboard

Abstract

The application discloses a method for generating a video story outline, and belongs to the technical field of big data and video production. The method comprises the following steps: constructing an algorithm model through data; inputting a video introduction into the algorithm model as algorithm model parameters; inputting a keyword of a desired video story; obtaining a key set from the model according to the keyword; inputting the key set into the algorithm model; and outputting a story outline with a specified number of words. The method can reduce the cost of video production by applying the algorithm model to the field of video production and assisting in generating a video story outline.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of big data and video production, in particular to a method for generating a video story outline. BACKGROUND

[0002] With the development of Internet technology and the improvement of network transmission rate, video has become the main way for people to obtain information in daily life. Traditional video story outline writing relies on the accumulation and inspiration of authors, which has great limitations. Moreover, traditional video production has a long production cycle and high cost. Therefore, it is urgent to reduce the cost of video production at each link. SUMMARY

[0003] The technical task of the present application is to provide a method for generating a video story outline, which uses an algorithm model and relies on a large amount of data as model parameters to assist in generating a story outline and reduce the cost of video production.

[0004] The technical solution adopted by the present application to solve its technical problems is:

[0005] A method for generating a video story outline, which constructs an algorithm model through data,

[0006] The algorithm model parameters are inputted in batches with a video introduction;

[0007] The keywords of the desired video story are inputted, and the key set is obtained from the model according to the keywords;

[0008] The key set is inputted into the algorithm model, and the story outline with a specified number of words is outputted.

[0009] This method constructs an algorithm model through a large amount of data, expands a piece of text according to the inputted keywords, and applies it to the field of video production to assist in generating a story outline.

[0010] Further, the specific implementation process of the method is as follows:

[0011] 1) Construct a video story outline data set, use a programming language to construct a list [], and insert the story outline data one by one;

[0012] 2) Cut the data in the story outline data set; traverse each piece of data in the story outline data set, and cut the data into a list composed of words;

[0013] 3) According to the inputted keywords, filter out the key data set: loop through the story outline data set, judge whether the keywords are in the list, if yes, put the list into the key data set list;

[0014] 4) Construct a model;

[0015] 5), according to the input of the keyword and the input of the text number of words, start the cycle from the model to take out the word, combine together, form the final story outline text.

[0016] Preferably, the model is built, each word in each data in the key data set is traversed, and it is judged whether the each word is in the model dictionary:

[0017] If a word is not in the model dictionary, a new dictionary is added to the model, the key is the word, and the value is a list, the list adds a dictionary, the key is the next word corresponding to the word, and the value is 1;

[0018] If the word is in the model dictionary, the dictionary in the list corresponding to the word is traversed, and it is judged whether there is the next word corresponding to the word; if not, a new dictionary is added to the list, the key is the next word corresponding to the word, and the value is 1; if yes, the value of the next word in the dictionary is added by 1.

[0019] Further, for the model {“word a”:[{“word b”:5}, {“word c”:2}], “word b”:[{“word a”:3}, {“word c”:2}], ……},

[0020] Each word in each data in the key data set is traversed, if the word is not in the model dictionary, it is added, word a is the current circulating word, word b is the next circulating word, then it is added as {“word a”:[{“word b”:1}], ……};

[0021] If yes, check whether there is a dictionary corresponding to the next word word c in the list corresponding to the word in the model dictionary, if not, add it: {“word a”:[{“word b”:1}, {“word c”:1}],}; if yes, add 1 to the corresponding value in the dictionary.

[0022] Preferably, according to the input of the keyword and the input of the text number of words, start the cycle from the model to take out the word,

[0023] An empty result list [] is established, and the keyword is put in;

[0024] According to the input of the keyword, a word selection number dictionary is built;

[0025] Assigning an initial value of 0, generating a random number k between 0 and 1 (the purpose of generating a random number is to increase the diversity of the results), if k < 0.5, taking the 0th dictionary in the keyword corresponding list from the model, taking its key and putting it into the result list; Then take out the word corresponding list from the model, check if the word selection number dictionary has this word;

[0026] If k > 0.5, take the 1st dictionary in the keyword corresponding list from the model, take its key and put it into the result list; Then take out the word corresponding list from the model, check if the word selection number dictionary has this word, if not, assign an initial value of 1, if yes, assign +1;

[0027] Until the number of elements in the result list reaches the specified number, then concatenate the result list into a string according to the order, which is the final story outline text output.

[0028] Preferably, the empty result list [] is established, and the keywords are put into the keyword corresponding list in the model, and the values in each dictionary in the model are sorted in descending order.

[0029] Preferably, the word selection number dictionary is constructed according to the input keywords, the key is the input keyword, and the value is 1.

[0030] Further, the specific implementation process of the loop taking out the word from the model is as follows:

[0031] Generating a random number k between 0 and 1, if k < 0.5, taking the 0th dictionary in the keyword corresponding list from the model, taking its key and putting it into the result list; If k > 0.5, take the 1st dictionary in the keyword corresponding list from the model, take its key and put it into the result list;

[0032] Taking the key as a new word, taking the word corresponding list from the model; Check if the word selection number dictionary has this word;

[0033] If not, add a new data to the word selection number dictionary, the key is the new word, the value is 1, and the key is taken and put into the result list; If yes, continue to execute: generating a random number k between 0 and 1, if k < 0.5, taking the value corresponding to the word selection number dictionary, setting it as variable n, taking the n th dictionary in the list, taking its key and putting it into the result list, changing the value corresponding to the word selection number dictionary to n+1; If k > 0.5, take the value corresponding to the word selection number dictionary, set it as variable n, take the n+1th dictionary in the list, take its key and put it into the result list, change the value corresponding to the word selection number dictionary to n+2;

[0034] According to the input text word number, a loop is started, and the number of loops is equal to the word number;

[0035] The words in the result list are spliced in order to form a final story outline text output.

[0036] The application also claims a device for generating a video story outline, comprising at least one memory and at least one processor;

[0037] The at least one memory is used to store a machine-readable program;

[0038] The at least one processor is used to call the machine-readable program to realize the method for generating a video story outline.

[0039] The application also claims a computer readable medium, wherein computer instructions are stored on the computer readable medium, and the computer instructions make the processor execute the method for generating a video story outline when the processor executes the computer instructions.

[0040] Compared with the prior art, the method for generating a video story outline has the following beneficial effects:

[0041] The method model is suitable for video story outline writing scenes and fields, and can generate a story outline by constructing an algorithm model through a large amount of data, expanding a paragraph of text according to input keywords, and applying the text to the video production field. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 The method for generating a video story outline provided by the embodiment of the application is shown in the flowchart. DETAILED DESCRIPTION

[0043] The application will be further described below in combination with specific embodiments.

[0044] The embodiment of the application provides a method for generating a video story outline, and an algorithm model is constructed through data.

[0045] The algorithm model parameters are batch input video introductions;

[0046] The keywords of a desired video story are input, and a key set is obtained from the model according to the keywords;

[0047] The key set is input into the algorithm model, and a story outline with a specified number of words is output.

[0048] The method constructs an algorithm model through a large amount of data, expands a paragraph of text according to input keywords, and applies the text to the video production field to assist in generating a story outline.

[0049] The specific implementation process of the method is as follows:

[0050] 1. Construct a video story outline data set.

[0051] Use the python programming language (not limited to one programming language, python, java or other) to construct a list [], and insert the story outline data one by one;

[0052] For example:

[0053] ["The film tells the story of the protagonist David Cook (Lakeland Woods) who is a data processing clerk in the far-tooth bank with a bleak future. In fact, he is also an artist. David loves comics, especially the super hero Fire Fox Kao, and dreams of meeting her creator, Toba. In order to get an opportunity to meet his idol, he decides to participate in a comic competition held in Japan. One day, he met the real "Fire Fox Kao" when he went to the bar with his colleague Gary Barrett (Andrew Ryan). The shoes she accidentally dropped when she left were picked up by David. This strange encounter left a deep impression on David, and at the same time, his life was quietly changing.",]

[0054] 2. Segment the data in the story outline data set.

[0055] Traverse each piece of data in the story outline data set and segment the data into a list of words; as follows:

[0056] ['The film tells the story of David, a data entry clerk at a bank who is also an artist. David loves comics, and is especially obsessed with the superhero Fox-Kaos. He dreams of meeting her creator, Shigeo Fujimoto. To get an opportunity to meet his idol, he decides to enter a manga contest in Japan. One night, he meets the real Fox-Kaos at a bar with his co-worker, Gary Barrett. Fox-Kaos drops her shoe, which is picked up by David. This strange encounter leaves a deep impression on David, and his life is slowly changing.']

[0057] 3. Filter out the key data set according to the input key word:

[0058] Loop through the story outline data set to determine if the key word is in the list. If it is, put the list into the key data set list.

[0059] 4) Build a model;

[0060] Loop through each word of each data in the key data set to determine if each word is in the model dictionary:

[0061] If a word is not in the model dictionary, add a new dictionary to the model, with the word as the key and a list as the value. Add a dictionary to the list, with the next word corresponding to the word as the key and 1 as the value.

[0062] If the word is in the model dictionary, traverse the list corresponding to the word to determine if there is a next word corresponding to the word; if not, add a new dictionary to the list with the next word as the key and 1 as the value; if so, add 1 to the value of the next word in the dictionary.

[0063] For example, {“word a”:[{“word b”:5}, {“word c”:2}], “word b”:[{“word a”:3}, {“word c”:2}], ……},

[0064] Traverse each word of each piece of data in the key data set. If the word is not in the model dictionary, add it. If the word a is the current word and the word b is the next word, add {“word a”:[{“word b”:1}], ……}.

[0065] If so, check if the word c corresponding to the next word in the list corresponding to the word in the model dictionary. If not, add {“word a”:[{“word b”:1}, {“word c”:1}],}. If so, add 1 to the corresponding value in the dictionary.

[0066] 5. According to the input keywords and the input text length, start the loop to take out the words from the model, combine them together, and form the final story outline text.

[0067] Create an empty result list [] and put the keywords in it. Sort the values in each dictionary in the model from large to small.

[0068] According to the input keywords, construct a word selection number dictionary; the key is the input keyword, and the value is 1.

[0069] Assign an initial value of 0 to generate a random number k between 0 and 1 (the purpose of generating a random number is to increase the diversity of the results). If k<0.5, take out the 0th dictionary in the list corresponding to the keyword from the model, take out its key and put it into the result list. Then take out the word list from the model and check if the word selection number dictionary has this word. If not, assign an initial value of 0. If so, assign +1 to it and take out its key and put it into the result list.

[0070] If k>0.5, take out the 1st dictionary in the list corresponding to the keyword from the model, take out its key and put it into the result list. Then take out the word list from the model and check if the word selection number dictionary has this word. If not, assign an initial value of 1. If so, assign +1 to it.

[0071] Until the number of result list elements reaches the specified number, then the result list is spliced into a string in order, which is the final story outline text output.

[0072] The embodiment gives the specific implementation process of the cycle taking out the word from the model, as shown in the following: Figure 1 The specific implementation process is as follows:

[0073] A random number k between 0 and 1 is generated, if k<0.5, the 0th dictionary in the list corresponding to the keyword is taken out from the model, and the key is taken out and put into the result list; if k>0.5, the 1st dictionary in the list corresponding to the keyword is taken out from the model, and the key is taken out and put into the result list;

[0074] The key taken out is taken as a new word, and the list corresponding to the word is taken out from the model; it is checked whether the word exists in the word selection number dictionary;

[0075] If not, a new data is added to the word selection number dictionary, the key is the new word, the value is 1, and the key is taken out and put into the result list; if yes, the following is executed: a random number k between 0 and 1 is generated, if k<0.5, the value corresponding to the word selection number dictionary is taken out and set as a variable n, the nth dictionary in the list is taken out, the key is taken out and put into the result list, and the value corresponding to the word selection number dictionary is changed to n+1; if k>0.5, the value corresponding to the word selection number dictionary is taken out and set as a variable n, the n+1th dictionary in the list is taken out, the key is taken out and put into the result list, and the value corresponding to the word selection number dictionary is changed to n+2;

[0076] The cycle starts according to the number of input text words, and the number of cycles is equal to the number of words;

[0077] The words in the result list are spliced in order to form the final story outline text output.

[0078] The embodiment of the application further provides a device for generating a video story outline, comprising at least one memory and at least one processor.

[0079] The at least one memory is used to store a machine readable program;

[0080] The at least one processor is used to call the machine readable program, and realize the method for generating a video story outline as described in the above embodiment.

[0081] The embodiment of the present application also provides a computer readable medium, which stores computer instructions, and the computer instructions make the processor execute the method for generating the video story outline in the above embodiment when executed by the processor. Specifically, a system or device equipped with a storage medium can be provided, and the storage medium stores software program codes for realizing the functions of any one of the above embodiments, and the computer (or CPU or MPU) of the system or device reads and executes the program codes stored in the storage medium.

[0082] In this case, the program codes read from the storage medium can realize the functions of any one of the above embodiments, and thus the program codes and the storage medium storing the program codes constitute a part of the present application.

[0083] The storage medium for providing the program codes includes a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program codes can be downloaded from a server computer through a communication network.

[0084] In addition, it should be clear that not only the program codes read by the computer can be executed, but also part or all of the actual operations can be completed by the operating system or the like operating on the computer based on the instructions of the program codes, so as to realize the functions of any one of the above embodiments.

[0085] In addition, it should be understood that the program codes read from the storage medium can be written into the memory provided in the expansion board inserted into the computer or the memory provided in the expansion unit connected to the computer, and then part or all of the actual operations can be executed by the CPU or the like installed on the expansion board or the expansion unit based on the instructions of the program codes, so as to realize the functions of any one of the above embodiments.

[0086] The present application has been described in detail by the above drawings and preferred embodiments, however, the present application is not limited to the disclosed embodiments, and those skilled in the art can know that the code review means in the above different embodiments can be combined to obtain more embodiments of the present application, and these embodiments are also within the protection scope of the present application.

Claims

1. A method for generating a video story outline, characterized in that: An algorithm model is constructed by data, wherein the algorithm model Video introduction to batch input of algorithm model parameters; Enter the keywords of the desired video story; obtain the key set from the model based on the keywords; Input the key set into the algorithm model; output the story outline of the specified number of words; The specific implementation process of this method is as follows: 1) Build a video story outline dataset, use programming language to build a list [], and insert the story outline data one by one; 2) Split the data in the story outline dataset; Traverse each data point in the story outline dataset and split the data into lists of words; 3) Filter out key datasets based on the input keywords: loop through the story outline dataset to determine whether the keywords are in the list. If so, add the list to the key dataset list; 4) Build the model; 5) Based on the input keywords and the number of words in the text, start looping to extract words from the model and combine them together to form the text of the final story outline; The model is constructed by traversing each word of each data in the key data set and determining whether each word is in the model dictionary: If a word is not in the model dictionary, a new dictionary is added to the model, with the key being the word and the value being a list. A dictionary is added to the list, with the key being the next word corresponding to the word and the value being 1. If the word is in the model dictionary, traverse the dictionary in the list corresponding to the word to determine whether there is a next word corresponding to the word; if not, add a new dictionary to the list with the key being the next word corresponding to the word and the value being 1; if so, add 1 to the value of the next word in the dictionary; According to the input keywords and the number of words in the input text, the loop starts to extract words from the model and combine them together. Create an empty result list [] and put in keywords; According to the input keywords, construct a word selection dictionary; The specific implementation process of the loop to extract words from the model is as follows: Generate a random number k between 0 and 1. If k < 0.5, take the dictionary at position 0 in the list corresponding to the keyword from the model, take out its key and put it into the result list; if k > 0.5, take the dictionary at position 1 in the list corresponding to the keyword from the model, take out its key and put it into the result list; Take the extracted key as a new word, and extract the corresponding list of words from the model; check whether the word selection dictionary has this word; If not, then add a new data to the word selection dictionary, the key is the new word, the value is 1, and the key is taken out and put into the result list; if yes, then continue to execute: generate a random number k between 0 and 1, if k<0.5, take out the value corresponding to the word selection dictionary, set it as variable n, take out the dictionary at the nth position in the list, take out its key and put it into the result list, and change the value corresponding to the word selection dictionary to n+1; if k>0.5, take out the value corresponding to the word selection dictionary, set it as variable n, take out the dictionary at the n+1th position in the list, take out its key and put it into the result list, and change the value corresponding to the word selection dictionary to n+2; Start looping according to the number of words in the text input, and the number of loops is equal to the number of words; Concatenate the words in the result list in order to form the text output of the final story outline.

2. The method for generating a video story outline according to claim 1, wherein: For the model {"word a":[{"word b":5},{"word c":2}], "word b":[{"word a":3},{"word c":2}], ...}, Traverse each word of each data in the key data set. If the word is not in the model dictionary, add it. If word a is the word of the current loop and word b is the next word of the current loop, then add {"word a":[{"word b":1}],...}; If so, check whether there is a dictionary corresponding to the next word c in the list corresponding to this word in the model dictionary. If not, add it in: {"word a":[{"word b":1},{"word c":1}],}; if so, add 1 to the corresponding value in the dictionary.

3. The method for generating a video story outline according to claim 1, wherein: The method creates an empty result list [], puts in keywords, and sorts the values ​​in each dictionary in the model from large to small.

4. The method for generating a video story outline according to claim 1, wherein: The word selection dictionary is constructed based on the input keyword, the key is the input keyword, and the value is 1.

5. A device for generating a video story outline, characterized in that: include: at least one memory and at least one processor; The at least one memory is configured to store a machine-readable program; The at least one processor is configured to call the machine-readable program to implement the method according to any one of claims 1 to 4.

6. A computer-readable medium, characterized in that The computer readable medium stores computer instructions, which, when executed by a processor, enable the processor to perform the method according to any one of claims 1 to 4.

Images