A video generation method and apparatus

Abstract

The present invention discloses a video generation method and device, which includes: obtaining video demand information; the video demand information is related to a target product; processing the video demand information using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a plurality of hot videos to be used; processing the set of hot videos to be used to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product. It can be seen that the present invention can use the hot video processing rule to process the video demand information to obtain hot videos to be used, and then obtain the target creative video used for marketing the target product through comprehensive processing of the hot videos to be used, which is conducive to improving video generation efficiency and video quality, thereby reducing marketing costs.

Description

Technical Field

[0001] The present invention relates to the field of video technology, and in particular to a video generation method and device. Background Art

[0002] Marketing creative generation requires transitioning from text to images and then to video. Currently, video generation typically relies on manual processing, connecting and adjusting multimodal modules. This results in relatively low video generation efficiency. Therefore, it is crucial to provide a video generation method and device to improve video generation efficiency and quality, thereby reducing marketing costs. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a video generation method and device, which can use the hot video processing rules to process the video demand information to obtain the hot video to be used, and then obtain the target creative video for marketing the target product through comprehensive processing of the hot video to be used, which is conducive to improving the video generation efficiency and video quality, and thus reducing marketing costs.

[0004] In order to solve the above technical problems, a first aspect of an embodiment of the present invention discloses a video generation method, the method comprising:

[0005] Acquiring video demand information; wherein the video demand information is related to the target product;

[0006] The video demand information is processed using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a plurality of hot videos to be used;

[0007] The set of standby hot videos is processed to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product.

[0008] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the video demand information includes product image information;

[0009] The method of processing the video demand information using a preset hot video processing rule to obtain a hot video set to be used includes:

[0010] Using preset commodity item recognition rules to segment and identify the commodity image information to obtain commodity category information;

[0011] The product category information and the preset set of hot videos to be matched are processed to obtain a set of hot videos to be used; the set of hot videos to be matched includes a plurality of hot videos to be matched.

[0012] As an alternative implementation manner, in the first aspect of the embodiments of the present invention, the processing of the commodity category information and the preset set of hot videos to be matched to obtain a set of hot videos to be used includes:

[0013] For any hot video to be matched, determine whether the hot video to be matched matches the commodity category information to obtain a matching determination result;

[0014] When the matching determination result is yes, determine that the hot video to be matched is one of the hot videos to be used;

[0015] When the matching determination result is no, determine that the hot video to be matched is an alternative hot video;

[0016] Select a quantity of the alternative hot videos that match a preset category threshold from all the alternative hot videos as the hot videos to be used.

[0017] As an alternative implementation manner, in the first aspect of the embodiments of the present invention, the processing of the set of hot videos to be used to obtain a set of target creative videos includes:

[0018] Process the set of hot videos to be used using a preset video size processing rule to obtain a set of size hot videos; the set of size hot videos includes several size hot videos;

[0019] Process the set of size hot videos using a preset video subtitle processing rule to obtain a set of subtitle hot videos; the set of subtitle hot videos includes several subtitle hot videos;

[0020] Process the set of subtitle hot videos using a preset video background music processing rule to obtain a set of target creative videos; the set of target creative videos includes several target creative videos.

[0021] As an alternative implementation manner, in the first aspect of the embodiments of the present invention, the video demand information includes output video size information;

[0022] The processing of the set of hot videos to be used using a preset video size processing rule to obtain a set of size hot videos includes:

[0023] Determine whether the output video size information exists in the video demand information to obtain a size determination result;

[0024] When the size determination result is yes, perform proportional scaling on the image main body area of the target commodity corresponding to the video demand information according to the output video size information to obtain a target commodity image;

[0025] For any to-be-used hotness video, perform equal-proportion scaling on the to-be-used hotness video according to the output video size information to obtain a scaled hotness video corresponding to the to-be-used hotness video;

[0026] Migrate the target commodity image to the scaled hotness video corresponding to the to-be-used hotness video to generate a size hotness video corresponding to the to-be-used hotness video.

[0027] As an optional implementation manner, in the first aspect of the embodiments of the present invention, the video demand information further includes publicity text information;

[0028] The processing of the size hotness video set by using a preset video subtitle processing rule to obtain a subtitle hotness video set includes:

[0029] Judge whether the publicity text information exists in the video demand information to obtain a text judgment result;

[0030] When the text judgment result is yes, use a preset subtitle image generation rule for the publicity text information to obtain a first input text image;

[0031] For any size hotness video, clear the subtitle information corresponding to the size hotness video to obtain a first subtitle-free hotness video corresponding to the size hotness video;

[0032] Migrate the first input text image to the first subtitle-free hotness video corresponding to the size hotness video to generate a subtitle hotness video corresponding to the size hotness video.

[0033] As an optional implementation manner, in the first aspect of the embodiments of the present invention, the video demand information further includes publicity recording information;

[0034] The processing of the subtitle hotness video set by using a preset video background music processing rule to obtain a target creative video set includes:<##

[0035] Judge whether the publicity recording information exists in the video demand information to obtain a recording judgment result;

[0036] When the recording judgment result is yes, for any subtitle hotness video, update the original background music corresponding to the subtitle hotness video by using the publicity recording information to obtain a target creative video corresponding to the subtitle hotness video;

[0037] When the recording judgment result is no, for any subtitle hotness video, screen out a target music material matching the target commodity word corresponding to the subtitle hotness video from a preset music material library as the publicity background music corresponding to the subtitle hotness video;

[0038] The original soundtrack corresponding to the subtitle hot video is updated using the promotional soundtrack corresponding to the subtitle hot video to obtain a target creative video corresponding to the subtitle hot video.

[0039] A second aspect of an embodiment of the present invention discloses a video generation device, comprising:

[0040] An acquisition module, configured to acquire video demand information related to a target product;

[0041] A first processing module is configured to process the video demand information using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a plurality of hot videos to be used;

[0042] The second processing module is used to process the standby hot video set to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product.

[0043] As an optional implementation manner, in the second aspect of the embodiment of the present invention, the video demand information includes product image information;

[0044] The first processing submodule includes an identification submodule and a processing submodule, wherein:

[0045] The identification submodule is used to segment and identify the product image information using a preset product identification rule to obtain product category information;

[0046] The processing submodule is used to process the product category information and the preset set of hot videos to be matched to obtain a set of hot videos to be used; the set of hot videos to be matched includes several hot videos to be matched.

[0047] As an optional implementation, in the second aspect of the embodiment of the present invention, the processing submodule processes the product category information and the preset hot video set to be matched to obtain the hot video set to be used in the following specific manner:

[0048] For any hot video to be matched, determine whether the hot video to be matched matches the product category information and obtain a matching judgment result;

[0049] When the matching judgment result is yes, determining that the hot video to be matched is one of the hot videos to be used;

[0050] When the matching judgment result is no, determining that the hot video to be matched is a candidate hot video;

[0051] Select a quantity of the alternative popular videos that match a preset category threshold from all the alternative popular videos as the standby popular videos.

[0052] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, the specific manner in which the second processing module processes the standby popular video set to obtain a target creative video set is as follows:

[0053] Process the standby popular video set using a preset video size processing rule to obtain a size popular video set; the size popular video set includes a number of size popular videos;

[0054] Process the size popular video set using a preset video subtitle processing rule to obtain a subtitle popular video set; the subtitle popular video set includes a number of subtitle popular videos;

[0055] Process the subtitle popular video set using a preset video background music processing rule to obtain a target creative video set; the target creative video set includes a number of target creative videos.

[0056] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, the video requirement information includes output video size information;

[0057] The standby popular video set further includes the image main area of the target commodity corresponding to the video requirement information;

[0058] The specific manner in which the second processing module processes the standby popular video set using a preset video size processing rule to obtain a size popular video set is as follows:

[0059] Determine whether the output video size information exists in the video requirement information to obtain a size determination result;

[0060] When the size determination result is yes, perform equal-proportion scaling on the image main area of the target commodity corresponding to the video requirement information according to the output video size information to obtain a target commodity image;

[0061] For any standby popular video, perform equal-proportion scaling on the standby popular video according to the output video size information to obtain a scaled popular video corresponding to the standby popular video;

[0062] Migrate the target commodity image to the scaled popular video corresponding to the standby popular video to generate a size popular video corresponding to the standby popular video.

[0063] As an alternative implementation manner, in the second aspect of the embodiments of the present invention, the video requirement information further includes promotional text information;

[0064] The specific manner in which the second processing module processes the size-heat video set by using a preset video subtitle processing rule to obtain a subtitle-heat video set is as follows:

[0065] Determine whether the promotional text information exists in the video requirement information to obtain a text judgment result;

[0066] When the text judgment result is yes, use a preset subtitle image generation rule for the promotional text information to obtain a first input text image;

[0067] For any size-heat video, clear the subtitle information corresponding to the size-heat video to obtain a first subtitle-free heat video corresponding to the size-heat video;

[0068] Migrate the first input text image to the first subtitle-free heat video corresponding to the size-heat video to generate a subtitle-heat video corresponding to the size-heat video.

[0069] As an optional implementation manner, in the second aspect of the embodiments of the present invention, the video requirement information further includes promotional audio information;

[0070] The specific manner in which the second processing module processes the subtitle-heat video set by using a preset video background music processing rule to obtain a target creative video set is as follows:

[0071] Determine whether the promotional audio information exists in the video requirement information to obtain an audio judgment result;

[0072] When the audio judgment result is yes, for any subtitle-heat video, update the original background music corresponding to the subtitle-heat video by using the promotional audio information to obtain a target creative video corresponding to the subtitle-heat video;

[0073] When the audio judgment result is no, for any subtitle-heat video, screen out a target music material matching the target commodity word corresponding to the subtitle-heat video from a preset music material library as the promotional background music corresponding to the subtitle-heat video;

[0074] Update the original background music corresponding to the subtitle-heat video by using the promotional background music corresponding to the subtitle-heat video to obtain a target creative video corresponding to the subtitle-heat video.

[0075] A third aspect of the present invention discloses another video generation device, and the device includes:

[0076] A memory storing executable program code;

[0077] A processor coupled to the memory;

[0078] The processor calls the executable program code stored in the memory to execute part or all of the steps in the video generation method disclosed in the first aspect of the embodiment of the present invention.

[0079] The fourth aspect of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all of the steps in the video generation method disclosed in the first aspect of the embodiment of the present invention.

[0080] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0081] In an embodiment of the present invention, video demand information is obtained; the video demand information is related to a target product; the video demand information is processed using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a number of hot videos to be used; the set of hot videos to be used is processed to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product. It can be seen that the present invention can use the hot video processing rule to process the video demand information to obtain hot videos to be used, and then obtain the target creative video for marketing the target product through comprehensive processing of the hot videos to be used, which is conducive to improving video generation efficiency and video quality, thereby reducing marketing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0083] Figure 1 This is a flow chart of a video generation method disclosed in an embodiment of the present invention;

[0084] Figure 2 is a flowchart of another video generation method disclosed in an embodiment of the present invention;

[0085] Figure 3 It is a structural diagram of a video generating device disclosed in an embodiment of the present invention;

[0086] Figure 4 is a structural diagram of another video generating device disclosed in an embodiment of the present invention;

[0087] Figure 5 A schematic structural diagram of another video generating device disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0088] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0089] The terms "first," "second," and so on, in the description and claims of the present invention and the accompanying drawings are used to distinguish between different objects, not to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or device.

[0090] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0091] The present invention discloses a video generation method and device that utilizes trending video processing rules to process video demand information to generate candidate trending videos. Furthermore, through comprehensive processing of the candidate trending videos, target creative videos for marketing target products are generated. This method is beneficial for improving video generation efficiency and video quality, thereby reducing marketing costs. Details are provided below.

[0092] Example 1

[0093] See also Figure 1 , Figure 1 This is a flow chart of a video generation method disclosed in an embodiment of the present invention. Figure 1 The video generation method described is applied to a video management system, such as a local server or cloud server for video generation management, etc., which is not limited in the embodiment of the present invention. Figure 1 As shown, the video generation method may include the following operations:

[0094] 101. Obtain video demand information.

[0095] In the embodiment of the present invention, the video demand information is related to the target product.

[0096] 102. Use a preset hot video processing rule to process the video demand information to obtain a set of hot videos to be used.

[0097] In the embodiment of the present invention, the above-mentioned unused hot video set includes a plurality of unused hot videos.

[0098] 103. Process the target popular video set to obtain a target creative video set.

[0099] In the embodiment of the present invention, the target creative video set is used to indicate the marketing of the target product.

[0100] Optionally, the above-mentioned video demand information includes product image information, and / or output video size information, and / or promotional text information, and / or promotional recording information, which is not limited in the embodiment of the present invention.

[0101] Optionally, the above-mentioned product image information includes one or more target product images.

[0102] Optionally, the above-mentioned hot video collection to be used also includes product category information.

[0103] Optionally, the above-mentioned product category information includes the image main area of the target product corresponding to the video demand information.

[0104] It can be seen that the video generation method described in the embodiment of the present invention can use the hot video processing rules to process the video demand information to obtain the hot video to be used, and then obtain the target creative video for marketing the target product through comprehensive processing of the hot video to be used, which is conducive to improving the video generation efficiency and video quality, and thus reducing marketing costs.

[0105] In an optional embodiment, the video demand information includes product image information;

[0106] In the above step 102, the video demand information is processed using the preset hot video processing rules to obtain a set of hot videos to be used, including:

[0107] Use the preset commodity item recognition rules to segment and identify the commodity image information to obtain commodity category information;

[0108] The product category information and the preset set of hot videos to be matched are processed to obtain a set of hot videos to be used; the set of hot videos to be matched includes several hot videos to be matched.

[0109] Optionally, the above-mentioned product category information includes product main area information and / or target category information, which is not limited in this embodiment of the present invention.

[0110] In this optional embodiment, as an optional implementation manner, the specific method for segmenting and recognizing the commodity image information by using the preset single-commodity recognition rule to obtain the commodity category information is as follows:

[0111] Input the target commodity image in the above commodity image information into an image segmentation model to obtain the commodity main body area information;

[0112] Input the target commodity image in the above commodity image information into a single-commodity recognition model to obtain the target category information.

[0113] Optionally, the above commodity main body area information includes the image main body area of the target commodity corresponding to the video demand information.

[0114] It can be seen that implementing the video generation method described in the embodiments of the present invention can segment and recognize the commodity image information by using the single-commodity recognition rule to obtain the commodity category information, and then screen out several to-be-matched popularity videos from the to-be-matched popularity videos, which is beneficial to improving the video generation efficiency and video quality, and further reducing the marketing cost.

[0115] In another optional embodiment, the above processing of the commodity category information and the preset set of to-be-matched popularity videos to obtain the set of to-be-used popularity videos includes:

[0116] For any to-be-matched popularity video, determine whether the to-be-matched popularity video matches the commodity category information to obtain a matching judgment result;

[0117] When the matching judgment result is yes, determine that the to-be-matched popularity video is a to-be-used popularity video;

[0118] When the matching judgment result is no, determine that the to-be-matched popularity video is an alternative popularity video;

[0119] Screen out the number of alternative popularity videos that match the preset category threshold from all alternative popularity videos as the to-be-used popularity videos.

[0120] In this optional embodiment, as an optional implementation manner, the specific generation method of the above to-be-matched popularity video is as follows:

[0121] For any popular commodity, obtain the set of carousel popularity videos corresponding to the popular commodity;

[0122] For any carousel popularity video, use a detection model based on a video frame-based screen detection algorithm to detect the carousel popularity video to obtain key frame index information; the above key frame index information includes a plurality of key frame indexes; the above key frame indexes are related to the video frames corresponding to the video screen switching of the carousel popularity video;

[0123] Segment the carousel popularity video by using the above key frame index information to obtain a set of segmented videos corresponding to the carousel popularity video; the above set of segmented videos includes a number of segmented videos;

[0124] Perform extraction processing on the above set of segmented videos to obtain a set of frame images; the above set of frame images includes a number of frame images; the above frame images are the K-th frame images of the segmented videos; the above K is a positive integer greater than or equal to 1;

[0125] Process the above set of frame images by using an image segmentation model to obtain a set of segmented commodity main body images; the above set of segmented commodity main body images includes a number of segmented commodity main body images; each of the above segmented commodity main body images corresponds to a unique frame image;

[0126] For any segmented video, process the segmented video by using an image extraction model based on an image matching and positioning algorithm and the segmented commodity main body image corresponding to the segmented video to obtain a set of commodity main body region information corresponding to the segmented video; the above set of commodity main body region information includes a number of segmented commodity main body region information; the above segmented commodity main body region information includes a segmented commodity main body region and the main body center coordinates corresponding to the segmented commodity main body region;

[0127] For any segmented commodity main body region information, fill the segmented commodity main body region corresponding to the segmented commodity main body region information by using the surrounding pixel values corresponding to the segmented commodity main body region information to obtain updated commodity main body region information corresponding to the segmented commodity main body region information; the above surrounding pixel values are pixel values within a certain range around the segmented commodity main body region;

[0128] Use the above updated commodity main body region information to splice and update the segmented video corresponding to the segmented commodity main body region information to obtain a spliced template video;

[0129] For any spliced template video, process the subtitles of the spliced template video by using a subtitle detection model based on a preset video subtitle detection algorithm to obtain spliced subtitle information corresponding to the spliced template video; the above spliced subtitle information includes spliced subtitle text information, and / or, spliced subtitle region coordinate information; the above spliced subtitle information is used to construct a set of subtitle template information;

[0130] Input the splicing subtitle information corresponding to the splicing template video into a preset product text recognition model to obtain a set of splicing product word information corresponding to the splicing template video; the above-mentioned set of splicing product word information includes several pieces of splicing product word information; the above-mentioned splicing product word information includes splicing product words and / or splicing position information; the above-mentioned splicing position information is the position of the above-mentioned splicing product word in the splicing subtitle text information; the above-mentioned splicing product word information is used to construct a set of product word information;

[0131] Separate the splicing background music corresponding to the splicing template video to obtain splicing background music information; the above-mentioned splicing background music information includes splicing background music and / or background music index information; the above-mentioned splicing background music information is used to construct a set of background music information;

[0132] Process the splicing template video, the segmented product main body area information corresponding to the splicing template video, the subtitle template information set, the product word information set, and the background music information set to obtain the to-be-matched popularity video corresponding to this carousel popularity video.

[0133] It can be seen that implementing the video generation method described in the embodiments of the present invention can use different processing methods for the matching of the to-be-matched popularity video and the product category information to determine the to-be-used popularity video, which is beneficial to improving the video generation efficiency and video quality, and thus reducing the marketing cost.

[0134] Embodiment 2

[0135] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of another video generation method disclosed in the embodiments of the present invention. Among them, Figure 2 the described video generation method is applied in a video management system, such as a local server or a cloud server for video generation management, etc., which is not limited in the embodiments of the present invention. As Figure 2 shown, the video generation method may include the following operations:

[0136] 201. Obtain video requirement information.

[0137] 202. Process the video requirement information using a preset popularity video processing rule to obtain a set of to-be-used popularity videos.

[0138] 203. Process the set of to-be-used popularity videos using a preset video size processing rule to obtain a set of size popularity videos.

[0139] In the embodiments of the present invention, the above-mentioned set of size popularity videos includes several size popularity videos.

[0140] 204. Process the set of size popularity videos using a preset video subtitle processing rule to obtain a set of subtitle popularity videos.

[0141] In the embodiment of the present invention, the above-mentioned subtitle hot video set includes a plurality of subtitle hot videos.

[0142] 205. Process the subtitle popularity video set using a preset video soundtrack processing rule to obtain a target creative video set.

[0143] In an embodiment of the present invention, the target creative video set includes several target creative videos.

[0144] In the embodiment of the present invention, for the specific technical details and technical terminology of steps 201 and 202, reference can be made to the detailed description of steps 101 and 102 in the first embodiment, which will not be repeated in detail in the embodiment of the present invention.

[0145] It can be seen that the video generation method described in the embodiment of the present invention can use the hot video processing rules to process the video demand information to obtain the hot video to be used, and then use the video size processing rules, video subtitle processing rules and video music processing rules to comprehensively process the hot video to be used to obtain the target creative video for marketing the target product, which is conducive to improving the video generation efficiency and video quality, and thus reducing marketing costs.

[0146] In an optional embodiment, the video requirement information includes output video size information;

[0147] The set of hot videos to be used also includes the image main area of the target product corresponding to the video demand information;

[0148] In the above step 203, the preset video size processing rules are used to process the target popularity video set to obtain the size popularity video set, including:

[0149] Determine whether the video requirement information contains output video size information, and obtain a size determination result;

[0150] When the size judgment result is yes, the image main area of the target product corresponding to the video requirement information is scaled proportionally according to the output video size information to obtain the target product image;

[0151] For any standby hot video, scale the standby hot video in equal proportion according to the output video size information to obtain a scaled hot video corresponding to the standby hot video;

[0152] The target product image is migrated to the scaled hot video corresponding to the standby hot video, and a size hot video corresponding to the standby hot video is generated.

[0153] In this optional embodiment, as an alternative implementation, when the size judgment result is negative, the main image area of the target product corresponding to the video demand information is scaled proportionally according to the preset video scaling size information to obtain an alternative scaled product image;

[0154] For any standby popularity video, it is scaled proportionally according to the video scaling size information to obtain an alternative scaled popularity video corresponding to the standby popularity video;

[0155] The alternative scaled product image is migrated to the alternative scaled popularity video corresponding to the standby popularity video to generate a size-popularity video corresponding to the standby popularity video.

[0156] It can be seen that implementing the video generation method described in the embodiments of the present invention can select an appropriate processing method to process the standby popularity video based on the judgment of the video size information to obtain a size-popularity video, which is more conducive to improving the video generation efficiency and video quality, and thus reducing the marketing cost.

[0157] In another optional embodiment, the above video demand information further includes promotional text information;

[0158] The above-mentioned processing of the size-popularity video set using the preset video subtitle processing rules to obtain a subtitle-popularity video set includes:

[0159] Judge whether there is promotional text information in the video demand information to obtain a text judgment result;

[0160] When the text judgment result is positive, the promotional text information is used to generate a first input text image using the preset subtitle image generation rules;

[0161] For any size-popularity video, the subtitle information corresponding to the size-popularity video is cleared to obtain a first subtitle-free popularity video corresponding to the size-popularity video;

[0162] The first input text image is migrated to the first subtitle-free popularity video corresponding to the size-popularity video to generate a subtitle-popularity video corresponding to the size-popularity video.

[0163] In this optional embodiment, as an alternative implementation, the specific method of clearing the subtitle information corresponding to the size-popularity video to obtain a subtitle-free popularity video corresponding to the size-popularity video is:

[0164] Calculate the coordinates of the subtitle area corresponding to the size-popularity video to obtain the subtitle coordinate information corresponding to the size-popularity video;

[0165] According to the subtitle coordinate information corresponding to the video of this size and heat, clear the subtitle information corresponding to the video of this size and heat to obtain a subtitle-free video of this size and heat.

[0166] In this alternative embodiment, as another alternative implementation, when the text judgment result is negative, for any video of a certain size and heat, according to the product category information corresponding to the video of this size and heat, identify the size product word corresponding to the video of this size and heat;

[0167] Use the size product word corresponding to the video of this size and heat to replace and update the subtitle product word in the subtitle text information corresponding to the video of this size and heat to obtain the size subtitle information corresponding to the video of this size and heat;

[0168] Use the preset subtitle image generation rule for the size subtitle information corresponding to the video of this size and heat to obtain the second input text image;

[0169] Clear the subtitle information corresponding to the video of this size and heat to obtain the second subtitle-free video corresponding to the video of this size and heat;

[0170] Transfer the second input text image to the second subtitle-free video corresponding to the video of this size and heat to generate a subtitle and heat video corresponding to the video of this size and heat.

[0171] It can be seen that implementing the video generation method described in the embodiments of the present invention can select an appropriate processing method to process the video of a certain size and heat based on the judgment of the promotional text information to obtain a subtitle and heat video, which is more conducive to improving the video generation efficiency and video quality, and thus reducing the marketing cost.

[0172] In yet another alternative embodiment, the above video requirement information further includes promotional recording information;

[0173] The above-mentioned processing of the subtitle and heat video set using the preset video background music processing rule to obtain the target creative video set includes:

[0174] Judge whether there is promotional recording information in the video requirement information to obtain a recording judgment result;

[0175] When the recording judgment result is positive, for any subtitle and heat video, use the promotional recording information to update the original background music corresponding to the subtitle and heat video to obtain the target creative video corresponding to the subtitle and heat video;

[0176] When the recording judgment result is negative, for any subtitle and heat video, screen out the target music material that matches the target product word corresponding to the subtitle and heat video from the preset music material library as the promotional background music corresponding to the subtitle and heat video;

[0177] The original soundtrack corresponding to the subtitle hot video is updated using the promotional soundtrack corresponding to the subtitle hot video to obtain a target creative video corresponding to the subtitle hot video.

[0178] Optionally, the above-mentioned music material library includes several music materials.

[0179] It can be seen that the video generation method described in the embodiment of the present invention can be implemented to select an appropriate processing method to process the subtitle popularity video based on the judgment of the promotional recording information to obtain the target creative video, which is more conducive to improving video generation efficiency and video quality, thereby reducing marketing costs.

[0180] Example 3

[0181] See also Figure 3 , Figure 3 This is a schematic diagram of the structure of a video generation device disclosed in an embodiment of the present invention. Figure 3 The described device can be applied to a video management system, such as a local server or cloud server for video generation management, etc., which is not limited in the embodiment of the present invention. Figure 3 As shown, the device may include:

[0182] The acquisition module 301 is used to acquire video demand information; the video demand information is related to the target product;

[0183] The first processing module 302 is configured to process the video demand information using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a plurality of hot videos to be used;

[0184] The second processing module 303 is used to process the target hot video set to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product.

[0185] It can be seen that implementation Figure 3 The described video generation device can use the hot video processing rules to process the video demand information to obtain the hot videos to be used, and then obtain the target creative video for marketing the target product through comprehensive processing of the hot videos to be used, which is conducive to improving the video generation efficiency and video quality, and thus reducing marketing costs.

[0186] In another optional embodiment, Figure 4 As shown, the video demand information includes product image information;

[0187] The first processing submodule 302 includes an identification submodule 3021 and a processing submodule 3022, wherein:

[0188] An identification sub-module, configured to perform segmentation and identification on the commodity image information by using a preset commodity single-item identification rule, so as to obtain commodity category information;

[0189] A processing sub-module, configured to process the commodity category information and a preset set of to-be-matched popularity videos to obtain a set of to-be-used popularity videos; the set of to-be-matched popularity videos includes a plurality of to-be-matched popularity videos.

[0190] It can be seen that implementing Figure 4 the described video generation device, which performs segmentation and identification on the commodity image information by using the commodity single-item identification rule to obtain commodity category information, and then screens out a plurality of to-be-matched popularity videos from the to-be-matched popularity videos, is beneficial to improving the video generation efficiency and video quality, and further reducing the marketing cost.

[0191] In another alternative embodiment, as Figure 4 shown, the specific manner in which the processing sub-module 3022 processes the commodity category information and the preset set of to-be-matched popularity videos to obtain the set of to-be-used popularity videos is as follows:

[0192] For any to-be-matched popularity video, determine whether the to-be-matched popularity video matches the commodity category information to obtain a matching judgment result;

[0193] When the matching judgment result is yes, determine that the to-be-matched popularity video is a to-be-used popularity video;

[0194] When the matching judgment result is no, determine that the to-be-matched popularity video is an alternative popularity video;

[0195] Screen out a quantity of alternative popularity videos that match a preset category threshold from all the alternative popularity videos as the to-be-used popularity videos.

[0196] It can be seen that implementing Figure 4 the described video generation device, which selects different processing methods for the matching of the to-be-matched popularity videos and the commodity category information to determine the to-be-used popularity videos, is beneficial to improving the video generation efficiency and video quality, and further reducing the marketing cost.

[0197] In another alternative embodiment, as Figure 4 shown, the specific manner in which the second processing module 303 processes the set of to-be-used popularity videos to obtain the set of target creative videos is as follows:

[0198] Process the set of to-be-used popularity videos by using a preset video size processing rule to obtain a set of size popularity videos; the set of size popularity videos includes a plurality of size popularity videos;

[0199] Process the set of videos with size popularity using preset video subtitle processing rules to obtain a set of videos with subtitle popularity; the set of videos with subtitle popularity includes several videos with subtitle popularity.

[0200] Process the set of videos with subtitle popularity using preset video background music processing rules to obtain a set of target creative videos; the set of target creative videos includes several target creative videos.

[0201] It can be seen that implementing Figure 4 the described video generation device can process video demand information using popularity video processing rules to obtain standby popularity videos, and then comprehensively process the standby popularity videos through video size processing rules, video subtitle processing rules, and video background music processing rules to obtain target creative videos for marketing the target product, which is beneficial to improving video generation efficiency and video quality, and thus reducing marketing costs.

[0202] In another optional embodiment, as Figure 4 shown, the video demand information includes output video size information;

[0203] The specific way for the second processing module 303 to process the set of standby popularity videos using preset video size processing rules to obtain a set of videos with size popularity is as follows:

[0204] Judge whether there is output video size information in the video demand information to obtain a size judgment result;

[0205] When the size judgment result is yes, scale the image main area of the target product corresponding to the video demand information proportionally to obtain a target product image;

[0206] For any standby popularity video, scale the standby popularity video proportionally according to the output video size information to obtain a scaled popularity video corresponding to the standby popularity video;

[0207] Migrate the target product image to the scaled popularity video corresponding to the standby popularity video to generate a video with size popularity corresponding to the standby popularity video.

[0208] It can be seen that implementing Figure 4 the described video generation device can select an appropriate processing method to process the standby popularity videos by judging the video size information to obtain videos with size popularity, which is more beneficial to improving video generation efficiency and video quality, and thus reducing marketing costs.

[0209] In another optional embodiment, as Figure 4 shown, the video demand information further includes promotional text information;

[0210] The set of videos with standby heat also includes the image main area of the target commodity corresponding to the video demand information;

[0211] The specific way for the second processing module 303 to process the set of videos with dimension heat by using the preset video subtitle processing rules to obtain the set of videos with subtitle heat is as follows:

[0212] Judge whether there is promotional text information in the video demand information to obtain a text judgment result;

[0213] When the text judgment result is yes, use the preset subtitle image generation rules for the promotional text information to obtain the first input text image;

[0214] For any video with dimension heat, clear the subtitle information corresponding to the video with dimension heat to obtain the first subtitle-free video with dimension heat corresponding to the video with dimension heat;

[0215] Migrate the first input text image to the first subtitle-free video with dimension heat corresponding to the video with dimension heat to generate the video with subtitle heat corresponding to the video with dimension heat.

[0216] It can be seen that implementing Figure 4 The described video generation device can select an appropriate processing method based on the judgment of the promotional text information to process the videos with dimension heat to obtain the videos with subtitle heat, which is more conducive to improving the video generation efficiency and video quality, and thus reducing the marketing cost.

[0217] In another optional embodiment, as Figure 4 shown, the video demand information further includes promotional recording information;

[0218] The specific way for the second processing module 303 to process the set of videos with subtitle heat by using the preset video background music processing rules to obtain the set of target creative videos is as follows:

[0219] Judge whether there is promotional recording information in the video demand information to obtain a recording judgment result;

[0220] When the recording judgment result is yes, for any video with subtitle heat, update the original background music corresponding to the video with subtitle heat by using the promotional recording information to obtain the target creative video corresponding to the video with subtitle heat;

[0221] When the recording judgment result is no, for any video with subtitle heat, screen out the target music material matching the target commodity word corresponding to the video with subtitle heat from the preset music material library as the promotional background music corresponding to the video with subtitle heat;

[0222] Update the original background music corresponding to the video with subtitle heat by using the promotional background music corresponding to the video with subtitle heat to obtain the target creative video corresponding to the video with subtitle heat.

[0223] It can be seen that implementing Figure 4 the described video generation device can select an appropriate processing method based on the judgment of the promotional recording information to process the subtitle heat video to obtain the target creative video, which is more conducive to improving the video generation efficiency and video quality, and thus reducing the marketing cost.

[0224] Embodiment 4

[0225] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of another video generation device disclosed in the embodiments of the present invention. Among them, Figure 5 the described device can be applied to a video management system, such as a local server or a cloud server for video generation management, etc., which is not limited in the embodiments of the present invention. As Figure 5 shown, the device may include:

[0226] A memory 401 storing executable program code;

[0227] A processor 402 coupled to the memory 401;

[0228] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the video generation method described in Embodiment 1 or Embodiment 2.

[0229] Embodiment 5

[0230] The embodiments of the present invention disclose a computer-readable storage medium, which stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps in the video generation method described in Embodiment 1 or Embodiment 2.

[0231] Embodiment 6

[0232] The embodiments of the present invention disclose a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps in the video generation method described in Embodiment 1 or Embodiment 2.

[0233] The device embodiments described above are only illustrative. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place, or may be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0234] Through the above specific descriptions of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disk memories, tape memories, or any other medium that can be used to carry or store data and is computer-readable.

[0235] Finally, it should be noted that: The video generation method and device disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, rather than limiting them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A video generation method, characterized in that, The method comprises: Acquiring video demand information; wherein the video demand information is related to the target product; The video demand information is processed using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a plurality of hot videos to be used; Processing the unused hot video set to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product, and the video demand information includes product image information; The method of processing the video demand information using a preset hot video processing rule to obtain a hot video set to be used includes: Using preset commodity item recognition rules to segment and identify the commodity image information to obtain commodity category information; Processing the product category information and the preset set of hot videos to be matched to obtain a set of hot videos to be used; the set of hot videos to be matched includes a plurality of hot videos to be matched; The method for generating the hot videos to be matched is as follows: For any hot-selling product, obtain the carousel popular video set corresponding to the hot-selling product; For any popular carousel video, a detection model based on a video frame detection algorithm is used to detect the popular carousel video to obtain key frame index information; the popular carousel video is segmented using the key frame index information to obtain a segmented video set corresponding to the popular carousel video; the segmented video set is extracted to obtain a frame image set; the frame image set is processed using an image segmentation model to obtain a segmented product main body image set; For any segmented video, the segmented video is processed using an image extraction model based on an image matching positioning algorithm and a segmented product main image corresponding to the segmented video to obtain a product main area information set corresponding to the segmented video, wherein the product main area information set includes a plurality of segmented product main area information; For any of the segmented product main area information, fill the segmented product main area corresponding to the segmented product main area information with surrounding pixel values corresponding to the segmented product main area information to obtain updated product main area information corresponding to the segmented product main area information; and use the updated product main area information to splice and update the segmented videos corresponding to the segmented product main area information to obtain a spliced template video; For any spliced template video, the subtitles corresponding to the spliced template video are processed using a preset subtitle detection model based on a video subtitle detection algorithm to obtain spliced subtitle information corresponding to the spliced template video, and the spliced subtitle information is used to construct a subtitle template information set; the spliced subtitle information corresponding to the spliced template video is input into a preset product text recognition model to obtain a spliced product word information set corresponding to the spliced template video, the spliced product word information set including a plurality of spliced product word information, and the spliced product word information is used to construct a product word information set; the spliced soundtrack corresponding to the spliced template video is separated to obtain spliced soundtrack information, and the spliced soundtrack information is used to construct a soundtrack information set; Process the spliced template video, the segmented product main body area information corresponding to the spliced template video, the subtitle template information set, the product word information set, and the background music information set to obtain the to-be-matched popularity video corresponding to the carousel popularity video.

2. The video generation method according to claim 1, wherein The processing of the product category information and the preset to-be-matched popularity video set to obtain the to-be-used popularity video set includes: For any to-be-matched popularity video, determine whether the to-be-matched popularity video matches the product category information to obtain a matching judgment result; When the matching judgment result is yes, determine the to-be-matched popularity video as one of the to-be-used popularity videos; When the matching judgment result is no, determine the to-be-matched popularity video as an alternative popularity video; Select the number of alternative popularity videos that match the preset category threshold from all the alternative popularity videos as the to-be-used popularity videos.

3. The video generation method according to claim 1, wherein The processing of the to-be-used popularity video set to obtain the target creative video set includes: Process the to-be-used popularity video set using the preset video size processing rule to obtain a size popularity video set; the size popularity video set includes several size popularity videos; Process the size popularity video set using the preset video subtitle processing rule to obtain a subtitle popularity video set; the subtitle popularity video set includes several subtitle popularity videos; Process the subtitle popularity video set using the preset video background music processing rule to obtain the target creative video set; the target creative video set includes several target creative videos.

4. The video generation method according to claim 3, wherein The video requirement information includes output video size information; The to-be-used popularity video set further includes the image main body area of the target product corresponding to the video requirement information; The processing of the to-be-used popularity video set using the preset video size processing rule to obtain a size popularity video set includes: Determine whether the output video size information exists in the video requirement information to obtain a size judgment result; When the size judgment result is yes, perform proportional scaling on the image main body area of the target product corresponding to the video requirement information according to the output video size information to obtain a target product image; For any to-be-used popularity video, perform proportional scaling on the to-be-used popularity video according to the output video size information to obtain a scaled popularity video corresponding to the to-be-used popularity video; Migrate the target product image to the scaled popularity video corresponding to the to-be-used popularity video to generate a size popularity video corresponding to the to-be-used popularity video.

5. The video generation method according to claim 3, wherein The video requirement information further includes promotional text information; The processing of the size popularity video set using the preset video subtitle processing rule to obtain a subtitle popularity video set includes: Determine whether the promotional text information exists in the video requirement information to obtain a text judgment result; When the text judgment result is yes, use the preset subtitle image generation rule for the promotional text information to obtain a first input text image; For any size of hot video, clear the subtitle information corresponding to the size of hot video to obtain the first wordless hot video corresponding to the size of hot video; The first input text image is migrated to a first wordless hot video corresponding to the hot video of the size, and a subtitle hot video corresponding to the hot video of the size is generated.

6. The video generation method according to claim 3, wherein The video demand information also includes promotional recording information; The method of processing the subtitle popularity video set using a preset video soundtrack processing rule to obtain a target creative video set includes: Determine whether the promotional recording information exists in the video demand information, and obtain a recording determination result; When the recording judgment result is yes, for any subtitle hot video, the original soundtrack corresponding to the subtitle hot video is updated using the promotional recording information to obtain a target creative video corresponding to the subtitle hot video; When the recording judgment result is no, for any subtitle hot video, a target music material matching the target product word corresponding to the subtitle hot video is selected from a preset music material library as the promotional soundtrack corresponding to the subtitle hot video; The original soundtrack corresponding to the subtitle hot video is updated using the promotional soundtrack corresponding to the subtitle hot video to obtain a target creative video corresponding to the subtitle hot video.

7. A video generation device, characterized in that, The device comprises: An acquisition module, configured to acquire video demand information related to a target product; A first processing module is configured to process the video demand information using a preset hot video processing rule to obtain a set of hot videos to be used; the set of hot videos to be used includes a plurality of hot videos to be used; a second processing module, configured to process the unused hot video set to obtain a target creative video set; the target creative video set is used to indicate the marketing of the target product, and the video demand information includes product image information; The first processing module includes an identification submodule and a processing submodule, wherein: The identification submodule is used to segment and identify the product image information using a preset product identification rule to obtain product category information; The processing submodule is used to process the product category information and the preset set of hot videos to be matched to obtain a set of hot videos to be used; the set of hot videos to be matched includes a plurality of hot videos to be matched; The method for generating the hot videos to be matched is as follows: For any hot-selling product, obtain the carousel popular video set corresponding to the hot-selling product; For any popular carousel video, a detection model based on a video frame detection algorithm is used to detect the popular carousel video to obtain key frame index information; the popular carousel video is segmented using the key frame index information to obtain a segmented video set corresponding to the popular carousel video; the segmented video set is extracted to obtain a frame image set; the frame image set is processed using an image segmentation model to obtain a segmented product main body image set; For any segmented video, use an image extraction model based on an image matching and positioning algorithm and the segmented commodity main body image corresponding to the segmented video to process the segmented video, and obtain a set of commodity main body region information corresponding to the segmented video. The set of commodity main body region information includes a number of segmented commodity main body region information; For any of the segmented commodity main body region information, use the surrounding pixel values corresponding to the segmented commodity main body region information to fill the segmented commodity main body region corresponding to the segmented commodity main body region information, and obtain updated commodity main body region information corresponding to the segmented commodity main body region information; use the updated commodity main body region information to splice and update the segmented video corresponding to the segmented commodity main body region information, and obtain a spliced template video; For any spliced template video, use a subtitle detection model based on a preset video subtitle detection algorithm to process the subtitle corresponding to the spliced template video, and obtain spliced subtitle information corresponding to the spliced template video. The spliced subtitle information is used to construct a set of subtitle template information; input the spliced subtitle information corresponding to the spliced template video into a preset commodity text recognition model, and obtain a set of spliced commodity word information corresponding to the spliced template video. The set of spliced commodity word information includes a number of spliced commodity word information, and the spliced commodity word information is used to construct a set of commodity word information; separate the spliced background music corresponding to the spliced template video to obtain spliced background music information, and the spliced background music information is used to construct a set of background music information; Process the spliced template video, the segmented commodity main body region information corresponding to the spliced template video, the set of subtitle template information, the set of commodity word information, and the set of background music information to obtain a to-be-matched hot video corresponding to the carousel hot video.

8. A video generation device, characterized in that, The device includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes the video generation method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer storage medium stores computer instructions, which are used to execute the video generation method according to any one of claims 1-6 when called.

Images