Poster generation method and device and electronic equipment

Abstract

The invention discloses a poster generation method and device and electronic equipment, and relates to the technical field of computers. The method comprises the following steps: receiving a poster generation demand, and constructing a demand structured vector based on the poster generation demand; performing template retrieval based on the demand structured vector to obtain a first poster template, and performing element adjustment on the first poster template based on an element adjustment instruction of the target user to obtain a second poster template; obtaining a plurality of target element contents corresponding to the second poster template; and performing element content replacement on the second poster template based on the multiple target element contents to obtain a target poster. The problem that a traditional poster generation method cannot meet personalized generation requirements of different users and cannot guarantee final quality and effect of poster generation is solved, personalized generation requirements of different users are met, and poster generation quality is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, in particular to a poster generation method and device and electronic equipment. BACKGROUND

[0002] As a visual communication tool, posters can quickly convey information, attract attention and evoke emotional resonance through the combination of graphics, text, color and other elements. With the advent of the digital age, the technology of automatically generating posters has emerged.

[0003] For poster generation, the existing technology usually uses existing poster templates to replace the content of poster elements. This method cannot meet the personalized needs of different users for poster generation, and the existing poster generation method lacks a detection method for poster generation, which cannot guarantee the final quality and effect of poster generation. SUMMARY

[0004] The main purpose of the present application is to provide a poster generation method, device and electronic equipment to solve the problem that the traditional poster generation method cannot meet the personalized generation needs of different users and cannot guarantee the final quality and effect of poster generation, which meets the personalized generation needs of different users and improves the quality of poster generation.

[0005] In order to achieve the above purpose, according to one aspect of the embodiments of the present application, a poster generation method is provided, which includes: receiving a poster generation requirement, and constructing a requirement structured vector based on the poster generation requirement; performing template retrieval based on the requirement structured vector to obtain a first poster template, and adjusting the elements of the first poster template based on the element adjustment instructions of the target user to obtain a second poster template; obtaining a plurality of target element contents corresponding to the second poster template; replacing the element contents of the second poster template based on the plurality of target element contents to obtain a target poster.

[0006] According to another aspect of the embodiments of the present application, a poster generation device is also provided, which includes: a construction unit configured to receive a poster generation requirement and construct a requirement structured vector based on the poster generation requirement; a retrieval unit configured to perform template retrieval based on the requirement structured vector to obtain a first poster template, and adjust the elements of the first poster template based on the element adjustment instructions of the target user to obtain a second poster template; an acquisition unit configured to obtain a plurality of target element contents corresponding to the second poster template; and a replacement unit configured to replace the element contents of the second poster template based on the plurality of target element contents to obtain a target poster.

[0007] According to still another aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores computer instructions for causing a computer to execute the above poster generation method.

[0008] According to a further aspect of the embodiments of the present application, an electronic device is also provided, which comprises at least one processor and a memory connected with the at least one processor in communication; the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the poster generation method.

[0009] The technical solutions provided by the embodiments of the present application can include the following beneficial effects: In the present application, the poster generation method, device and electronic device solve the problem that the traditional poster generation method cannot meet the personalized generation requirements of different users and cannot guarantee the final quality and effect of poster generation, meet the personalized generation requirements of different users, and improve the quality of poster generation. BRIEF DESCRIPTION OF DRAWINGS

[0010] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application, and make the other features, objectives and advantages of the present application more apparent. The illustrative embodiments of the present application and their description serve the purpose of explaining the present application. The accompanying drawings should not be regarded as a limitation of the present application. In the drawings: Figure 1 A schematic diagram of a hardware environment of an optional poster generation method provided by the present application; Figure 2 A flowchart of an optional poster generation method provided by the present application; Figure 3 A schematic diagram of an optional poster generation method provided by the present application; Figure 4 A schematic diagram of another optional poster generation method provided by the present application; Figure 5 A schematic diagram of a further optional poster generation method provided by the present application; Figure 6 A schematic diagram of a still further optional poster generation method provided by the present application; Figure 7 A structural schematic diagram of an optional poster generation device provided by the present application; Figure 8 A structural schematic diagram of an optional electronic device provided by the present application. DETAILED DESCRIPTION

[0011] In order to make the person skilled in the art better understand the scheme of the present application, the technical scheme in the embodiments of the present application will be clearly and completely described in the embodiments of the present application in combination with the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0012] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device. The embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0013] In order to solve the problem that the traditional poster generation method cannot meet the personalized generation needs of different users and cannot guarantee the final quality and effect of poster generation, the embodiments of the present application provide a poster generation method, device and electronic equipment. As an optional embodiment, the above-mentioned poster generation method can be applied to, but is not limited to, the poster generation system composed of a terminal device 102 and a server 104 as shown in Figure 1 As shown in Figure 1 The terminal device 102 is connected with the server 104 through a network 110. The network 110 can include but is not limited to a wired network and a wireless network. The wired network includes a local area network, a metropolitan area network and a wide area network. The wireless network includes Bluetooth, WIFI and other wireless communication networks. The terminal device 102 can include but is not limited to at least one of the following: a mobile phone (such as an Android mobile phone, an iOS mobile phone, etc.), a notebook computer, a tablet computer, a palm computer, a MID (Mobile Internet Devices), a PAD, a desktop computer, a smart television, a vehicle-mounted device, etc.

[0014] The terminal device 102 further includes a display 106, a processor 108, and a memory 112. The display 106 can be configured to display a poster generation process and a target poster. The processor 108 can be configured to process the collected data. The memory 112 can be configured to store relevant data. It can be understood that, when the terminal device 102 receives a poster generation request from a user, the terminal device 102 sends the poster generation request to the server 104 through the network 110, and the server 104 implements a specific poster generation process in response to the poster generation request.

[0015] The server 104 can be a single server, a server cluster composed of multiple servers, or a cloud server. The server 104 includes a database 114 and a processing engine 116. The database 114 can be configured to store data and models. The processing engine 116 can be configured to process the data.

[0016] According to an aspect of an embodiment of the present application, the poster generation system can further perform the following steps: first, the terminal device 102 performs S102 as shown in Figure 1 ; then, the server 104 performs S104 to S110 as shown in Figure 1 , thereby generating a target poster.

[0017] In the above embodiment of the present application, the poster generation method is used to solve the problem that the conventional poster generation method cannot meet the personalized generation requirements of different users and cannot guarantee the final quality and effect of poster generation, thereby meeting the personalized generation requirements of different users and improving the quality of poster generation.

[0018] As an optional implementation, refer to Figure 2 , which shows a flowchart of a poster generation method according to an embodiment of the present application. The method includes at least one of the following steps (S202 to S208): S202, receiving a poster generation requirement and constructing a requirement structured vector based on the poster generation requirement; S204, performing template retrieval based on the requirement structured vector to obtain a first poster template, and adjusting elements of the first poster template based on element adjustment instructions of a target user to obtain a second poster template; S206, obtaining a plurality of target element contents corresponding to the second poster template; and S208, replacing elements of the second poster template based on the plurality of target element contents to obtain a target poster.

[0019] In step S202 above, the poster generation requirement is used to instruct the generation of the poster needed by the target user. The requirement data carried in the poster generation requirement can be either voice data or text data. It is understood that if it is voice data, existing speech-to-text technology can be used to convert the voice data into text data. The specific text data can be set according to the specific personalized requirements of the target user. This application does not impose any restrictions. The process of constructing a requirement structure vector based on the poster generation requirement can be understood, but is not limited to, as being constructed based on the requirement data carried in the poster generation requirement. If the requirement data is text data, then the requirement structure vector is constructed based on the text data. If the requirement data is voice data, the requirement structure vector can be constructed based on the voice data, or the voice data can be converted into text data and then the requirement structure vector is constructed based on the text data.

[0020] In step S204 above, the poster to be labeled can be a poster template with preset content, or it can be a poster that needs to be labeled with elements for element replacement when generating a personalized poster later; each element information in the element information combination includes the element itself (also called element content) and the element's position, size, font and other attribute information in the poster.

[0021] The template retrieval in S204 above can be performed from a database storing poster templates, or from other storage locations storing poster templates. The first poster template retrieved is, for example, as shown below. Figure 3 The first poster template 300 shown (in the example poster template diagram shown in this application, the letters ABCDEFGHI represent the element content in the poster template, and the area occupied by the letters ABCDEFGHI represents the element area) allows the target user to further adjust the elements of the first poster template based on their personalized needs (i.e., the corresponding element adjustment instructions). The element adjustment in S204 is used to adjust the position information of the area occupied by the elements in the first poster template. The specific adjustment process includes: S1, obtaining the first element information of the first poster template, wherein the first element information includes the element content, the size of the element area, and the area position information corresponding to the size of the element area; S2, generating the first template element layer based on the element layer generation instructions and the first element information; S3, obtaining the element adjustment instructions and determining at least one element area to be adjusted in the first template element layer based on the element adjustment instructions; S4, adjusting the at least one element area to be adjusted according to the element adjustment instructions to obtain the second poster template.

[0022] The element layer generation instruction can be an instruction issued by the target user through mouse clicking. The element layer generation instruction can be a client imaging instruction or an aerial imaging instruction. The client imaging instruction is used to instruct generation of the first template element layer on the client. The aerial imaging instruction is used to instruct generation of the first template element layer in suspension in the air. In the case where the element layer generation instruction is the client imaging instruction, the first template element layer can be directly generated based on the first element information, and the first template element layer includes the element contents in the first element region, the element region size, and the region position information corresponding to the element region size. In this case, the first template element layer generation process can be, but is not limited to, understood as generating an element layer template that can be used by the target user to adjust the element region according to all information of the first poster template (i.e., the first element information), and the element contents, the element region size, and the region position information corresponding to the element region size in the first template element layer are the same as those of the first poster template. In the case where the element layer generation instruction is the aerial imaging instruction, generating the first template element layer based on the element layer generation instruction and the first element information includes specifically including: in the case where the element layer generation instruction instructs generation of the suspended template element layer, generating a reference template element layer based on the first element information, and obtaining a plurality of element regions in the reference template element layer; adaptively adding an identification region to each element region to obtain a plurality of reference element regions; and generating the first template element layer based on the plurality of reference element regions.

[0023] The above-mentioned adaptive addition of an identification region to each element region can be, but is not limited to, understood as adaptive addition of an identification region according to the element region size of each element region. Specifically, the identification region can be added according to the height information, the width information, and the specific position information of each element region. The region position of the identification region is adjacent to the region position of the element region, but the relative position between the identification region and the element region is not limited in the present application. For example, the identification region can be on the left side or the right side of the element region. The identification region can also be on the upper side or the lower side of the element region.

[0024] The above-mentioned adaptive addition of an identification region to each element region can be understood with reference to the example shown in FIG. 3A. As shown in FIG. 3A, the first poster template 300 includes an element region A, an element region B, an element region C, an element region D, an element region E, an element region F, an element region G, an element region H, and an element region I. An identification region is adaptively added to each element region in the first poster template. Figures 3-4 Figure 3 The above-mentioned adaptive addition of an identification region to each element region can be understood with reference to the example shown in FIG. 3A. As shown in FIG. 3A, the first poster template 300 includes an element region A, an element region B, an element region C, an element region D, an element region E, an element region F, an element region G, an element region H, and an element region I. An identification region is adaptively added to each element region in the first poster template. Figure 4 ​As shown, the identification area a is added to the element area A, the identification area b is added to the element area B, the identification area d is added to the element area D, the identification area e is added to the element area E, and the identification area I is added to the element area I. In order to ensure the display effect, this application only shows the part of adding the identification area to the element area located at the edge of the template in the first poster template 300, and does not show the effect of adding the identification area to the element areas C and G located in the middle, but it can be understood that this application adds the identification area to all element areas in the first poster template, which also includes adding the identification area to C and G.

[0025] The above-mentioned reference element area can be but is not limited to the area composed of the element area and the corresponding identification area, such as Figure 4 As shown, the element area A+identification area a constitutes a reference element area, the element area B+identification area b constitutes a reference element area, and the element area F+identification area f constitutes a reference element area. Then the element layer composed of multiple reference element areas is determined as the first template element layer. The identification area is used to facilitate the user to adjust the element in the floating template element layer (i.e. the first template element layer generated in the case that the element layer generation instruction indicates to generate the floating template element layer). In the case that the first template element layer is the floating template element layer, the user can realize the adjustment of the element area to be adjusted through the voice instruction of adjusting the identification area. At this time, the identification displayed in the identification area can be a red identification, which is convenient for the user to watch and issue instructions for adjustment.

[0026] The above-mentioned element adjustment instruction is the instruction corresponding to the element layer generation instruction. In the case that the element layer generation instruction indicates to generate the first template element layer in the client, the element adjustment instruction is the adjustment instruction corresponding to the first template element layer in the client (at this time, the element adjustment instruction can be the element adjustment instruction generated by the mouse click and drag operation of the target user), and in the case that the element layer generation instruction indicates to generate the floating instruction in the air (i.e. to generate the floating template element layer), the element instruction is the adjustment instruction corresponding to the floating template element layer in the air (at this time, the element adjustment instruction can be the adjustment instruction generated by analyzing the voice data or gesture data of the target user). According to the element adjustment instruction, at least one element area to be adjusted in the first template element layer which the target user wants to adjust can be determined, and then the element adjustment instruction is used to adjust at least one element area to be adjusted to obtain the second poster template.

[0027] Adjusting at least one element region to be adjusted according to element adjustment instructions (which can be understood as, but is not limited to, adjusting element regions) includes: performing a first element adjustment on at least one element region to be adjusted according to element adjustment instructions to obtain a reference poster template; obtaining the overall layout information of the first poster template (indicating the size of the first poster template) and the reference element information of the reference poster template; determining the relative position information of multiple reference element regions in the reference poster template based on the reference element information; and performing adaptive adjustments to the reference poster template based on the relative position information and the overall layout information to obtain a second poster template (which can be understood as, while ensuring that the size of the first poster template remains unchanged and the relative position information of the multiple reference element regions remains unchanged, adaptively adjusting the size of the multiple reference element regions). Specifically, this can be combined with... Figure 3 (It can also be) Figure 4 The first template element layer obtained after adding the identifier region is shown in the example in this application. Figure 3 (The first poster template 300 shown is used as an example to illustrate the concept.) Figures 5-6 understand: like Figure 5 As shown, suppose the target user needs to access information such as... Figure 3 The elements of the first template element layer corresponding to the first poster template 300 shown are adjusted, and the area of ​​elements to be adjusted determined based on the element adjustment instructions is as follows: Figure 3 The element regions B, D, and E shown are, after element adjustment, the result is as follows: Figure 5 The reference poster template 500 shown here (i.e., the actual element adjustment at this time involves swapping the positions of element area B with those of element areas D and E) clearly shows that the display effect of the element areas in the reference poster template is not suitable. Therefore, this application will perform regional adaptation adjustments on the reference poster template based on the first element information and the overall layout information to obtain a second poster template. The specific process is as follows: Figure 5 The reference poster template 500 shown has been adjusted as follows: Figure 6 The second poster template 600 shown is from... Figures 5 to 6 As can be seen from the regional adaptive adjustment effect, this application reduces the regional height of reference element regions F and G, increases the regional width of reference element regions D and E, and reduces the regional width and increases the regional height of reference element region B, thereby obtaining the following... Figure 6 The second poster template 600 is shown.

[0028] The obtaining of the plurality of target element contents in S206 can be, but is not limited to, obtaining a plurality of target element contents corresponding to a plurality of second element regions in the second poster template, and each element region corresponds to a target element content. The target element content can be obtained in the manner of obtaining the hot element content, such as through the TrendRadar open source tool to realize the automatic acquisition and update of hot data. The specific principle and process are as follows: the principle of hot acquisition is based on the technical scheme of “RSS subscription + LLM analysis”. TrendRadar integrates the RSS subscription interface of a plurality of mainstream news sources, crawls the hot information on the network in real time, and identifies the hot attributes (such as “holiday hotspots”, “social event hotspots”, “industry hotspots”, etc.) of the crawled content through LLM. Hot pushing and confirmation: the identified hot information is pushed to the enterprise chat tool (such as Dingding), and after the operation personnel click “confirm”, the hot spot is automatically updated to the “hot spot pool” (hot spot database), providing real-time hot spot materials for subsequent poster copy generation.

[0029] The above element content replacement of the second poster template based on the plurality of target element contents to obtain the target poster includes: obtaining a plurality of second element regions in the second poster target, and pairing the plurality of target element contents with the plurality of second element regions respectively to obtain a plurality of content-region pairs, wherein each target element content corresponds to a second element region; inputting the plurality of content-region pairs into a content adjustment model respectively to output a plurality of replacement content-region pairs; and replacing the element contents in the plurality of second element regions according to the plurality of replacement content-region pairs to obtain the target poster.

[0030] The above pairing can be, but is not limited to, pairing the target element content with the second element region, that is, determining which second element region the plurality of target element regions should be placed in.

[0031] And the above inputting the plurality of content-region pairs into the content adjustment model to output the plurality of replacement content-region pairs includes: determining any content-region pair as a current content-region pair, and obtaining region attribute information in the current content-region pair; performing semantic analysis on the content in the current content-region pair to obtain an analysis result; generating replacement content matched with the content in the current content-region pair based on the analysis result and the region attribute information, wherein the semantic similarity between the replacement content and the content in the current content-region pair is greater than a preset threshold; replacing the content in the current content-region pair with the replacement content to obtain a replacement content-region pair, and determining the content-region pairs other than the current content-region pair as the current content-region pair.

[0032] The specific process of inputting the plurality of content-region pairs into the content adjustment model respectively and outputting a plurality of replacement content-region pairs can be, but is not limited to, understood as processing each content-region pair respectively and processing it into a replacement content-region pair. The semantic analysis is used to determine the specific semantics of the content in the current content-region pair, and based on the analysis result and the region attribute information, the replacement content matching the content in the current content-region pair is generated. However, it can be, but is not limited to, understood that based on the analysis result and the region attribute information, the replacement content with a semantic similarity greater than or equal to a preset threshold to the content in the current content-region pair and a content (text) length adapted to the region attribute information is generated, and then the replacement content is replaced with the content in the current content-region, thereby obtaining a replacement content-region pair. For example, the original current content-region pair is a-A, and the generated replacement content is m, and the final replacement content-region pair obtained is m-A.

[0033] By the above-mentioned embodiments of the present application, the above-mentioned poster generation method is adopted, the problem that the traditional poster generation method cannot meet the personalized generation needs of different users and cannot guarantee the final quality and effect of poster generation is solved, the personalized generation needs of different users are met, and the poster generation quality is improved.

[0034] As an optional embodiment, the demand structured vector is constructed based on the poster generation demand, including: S1, keyword extraction is performed on the poster generation demand to obtain a plurality of poster keywords; S2, semantic analysis is performed on the poster generation demand to obtain a semantic analysis result; S3, demand structured data is constructed based on the plurality of poster keywords and the semantic analysis result; and S4, the demand structured data is converted into a demand structured vector.

[0035] For example, the plurality of poster keywords extracted in S1 are: happy report (poster type), wedding (application scenario of the poster), and perfect match (key text of the poster), respectively. The semantic analysis result obtained by performing semantic analysis on the poster is MNOPQ (the semantic analysis result is actually a text form analysis result, and the letters here are only an example). The demand structured data constructed in S3 is: happy report-wedding-perfect match-MNOPQ. Then, S4 is executed to convert the demand structured data into a demand structured vector.

[0036] It should be noted that the poster templates stored in the storage location (such as a database) in the present application each correspond to a corresponding poster structured vector and a corresponding poster template type, and the poster structured vector is determined based on the poster structured vector constructed based on the element content in the poster template, and the construction process of the poster structured vector includes: obtaining the poster element of each poster element region in the poster template, obtaining a plurality of poster elements, inputting the plurality of poster elements into an element semantic recognition model, and outputting a template element semantic (the semantic corresponding to the whole poster template); constructing template structured data based on the plurality of poster elements and the template element semantic; and converting the template structured data into a template structured vector. The construction method of the specific template structured data is the same as that of the demand structured data, and the conversion method of the specific template structured data into a template structured vector is also the same as that of the demand structured data into a demand structured vector. The conversion method of the structured data (demand structured data, template structured data) into the structured vector (demand structured vector, template structured vector) can adopt the existing method, which is not limited in the present application.

[0037] The above template retrieval based on the demand structured vector to obtain the first poster template specifically includes: determining the poster type according to the poster generation demand; determining a plurality of fourth poster templates from a plurality of third poster templates according to the poster type; obtaining reference template structured vectors corresponding to the plurality of fourth poster templates; determining the vector similarity between the demand structured vector and each reference template structured vector based on the demand structured vector; and determining the fourth poster template corresponding to the reference template structured vector corresponding to the highest vector similarity as the first poster template.

[0038] It should be noted that in the case where the fourth poster template corresponding to the reference template structured vector corresponding to the highest vector similarity is multiple, the first poster template can be determined from the multiple fourth poster templates corresponding to the reference template structured vector corresponding to the highest vector similarity based on the selection of the target user.

[0039] According to the above S1-S4 steps, the poster template closest to the user demand can be accurately selected from a plurality of poster templates.

[0040] As an optional implementation, after obtaining the target poster, it includes: S1, constructing detection context information according to the target poster, wherein the detection context information includes poster information and poster generation demand; S2, inputting the target poster and detection context information into a poster detection model to output a detection report and an optimization suggestion.

[0041] The S2 specifically includes: S2-1, determining a plurality of detection types corresponding to the target poster based on the detection context information; S2-2, performing poster detection on the target poster based on the detection context information and each detection type to obtain a detection result and an optimization suggestion corresponding to each detection type; and S2-3, generating a detection report based on the detection result corresponding to each detection type and generating an optimization suggestion report based on the optimization suggestion corresponding to each detection type.

[0042] The plurality of detection types include: a poster structure, a layout format, element content, and visual features, and the visual features include local visual features of each poster element and element region in the target poster and global visual features of the target poster.

[0043] The S2-2 to S2-3 specifically include, for example: detecting a poster structure of the target poster to obtain a first detection result and a first optimization suggestion; detecting a layout format of the target poster to obtain a second detection result and a second optimization suggestion; detecting element content of the target poster to obtain a third detection result and a third optimization suggestion; detecting visual features of the target poster to obtain a fourth detection result and a fourth optimization suggestion; and generating a detection report based on the first detection result, the second detection result, the third detection result, and the fourth detection result, and generating an optimization suggestion report corresponding to the detection report based on the first optimization suggestion, the second optimization suggestion, the third optimization suggestion, and the fourth optimization suggestion.

[0044] In the present application, the detection of the target poster after obtaining the target poster can adopt a specific poster detection system, and the poster detection system includes the following core layers: A data input layer for receiving poster data (containing information such as the position and content of a text element) and scene annotation rules.

[0045] An intelligent processing core layer including four core engines: a semantic annotation engine, a quality checking engine, an automated testing engine, and an AI quality detection engine.

[0046] A data support layer including a template library, an annotation knowledge base, a test case library, and a quality detection result library.

[0047] An AI capability layer including a large language model (for semantic understanding and annotation quality checking) and a visual understanding model (for quality detection).

[0048] In the present application, when detecting the target poster template, the following content (which can be understood as not limiting the specific detection steps) is specifically included: 1. Constructing a detection context: Obtain the generated poster rendering (i.e., the target poster); based on the poster rendering, construct detection context information, including poster information (poster title, poster annotation rate (annotation of each poster element using a poster template), total number of poster elements, number of annotated elements, number of replaced elements, etc.) and poster generation requirements; input the poster rendering and detection context information into the poster detection model (which can be an AI visual understanding model); 2. Poster detection model performs poster detection (which can be but not limited to AI visual understanding): call the poster detection model to intelligently identify the poster rendering, identify the text, layout, color, etc. visual elements in the poster rendering, identify the overall structure and visual hierarchy of the poster rendering, and analyze the visual features of the poster rendering; 3. Multi-dimensional evaluation of the poster: layout quality evaluation: check if the text layout is neat and beautiful, if the element alignment is reasonable, if the white space is appropriate, and if the overall layout is coordinated; replacement effect evaluation: check if the replaced content meets the user's requirements, if the semantic of the replaced content is accurate, if the visual effect after replacement is natural, and if there are any missing or incorrect replacements; overall effect evaluation: comprehensive evaluation of the overall visual effect, evaluation of whether it meets the design specifications, and evaluation of whether it achieves the expected effect; 4. Problem identification and classification: automatically identify problem types (layout problems, element replacement problems, etc.) and severity levels (low, medium, high), record the specific location and description of the problem; classify and prioritize the problems. (That is, after detecting problems based on multiple detection types, record the detection types and the corresponding detected problems (i.e., location and specific description), then classify the problems according to the detection types and sort the detected problems according to the severity of the problems.) 5. Optimization suggestion generation: generate specific optimization suggestions based on the detection results, provide actionable improvement plans, guide subsequent optimization work, and then optimize the target poster based on the optimization suggestions; 6. Detection result saving: save the detection results, including scores, problem lists, optimization suggestions, etc., generate a quality detection report, and support query and analysis of the detection results. That is, the application will ultimately generate a comprehensive quality detection report based on the detection report and the optimization suggestion report.

[0049] It should be noted that the model used in the present application is a model trained in advance using training samples. In the detection process of the target poster in the present application, the poster detection model can recognize and understand the visual content of the poster, as well as the text, layout, color and other elements. At the same time, based on the professional poster design standard, the layout quality and element replacement (i.e. the process of replacing the content with the replacement content) effect are quantitatively scored, and the problem type and severity are automatically identified, and specific optimization suggestions are generated based on the detection results. The poster detection model in the present application can also continuously optimize the generation quality of the poster based on the test results.

[0050] Through the above embodiments of the present application, the problem that the traditional poster generation method cannot meet the personalized generation needs of different users and cannot guarantee the final quality and effect of poster generation is solved, the personalized generation needs of different users are met, and the poster generation quality is improved.

[0051] In the present application, the detection of the target poster realizes objective evaluation from multiple dimensions such as layout quality, replacement effect and overall effect, automatically identifies poster problems and provides optimization suggestions. At the same time, through poster generation and detection, the whole process is automated, so that the poster generation, detection and other links are completed within a few seconds, the efficiency is greatly improved, and the poster quality detection algorithm inside the poster detection model is continuously optimized based on the verification and detection results, forming the continuous learning and optimization capability of the poster detection model.

[0052] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0053] According to another aspect of the embodiments of the present application, a poster generation device for implementing the above-mentioned poster generation method is also provided, as shown in Figure 7 The device comprises: a construction unit 702 configured to receive a poster generation requirement and construct a requirement structured vector based on the poster generation requirement; a retrieval unit 704 configured to perform template retrieval based on the requirement structured vector to obtain a first poster template, and perform element adjustment on the first poster template based on the element adjustment instruction of the target user to obtain a second poster template; an acquisition unit 706 configured to acquire a plurality of target element contents corresponding to the second poster template; a replacement unit 708 configured to replace the element contents of the second poster template based on the plurality of target element contents to obtain a target poster.

[0054] The specific manner in which each unit of the above device embodiment performs the operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0055] According to another aspect of the present invention, an electronic device for implementing the poster generation method is also provided, the electronic device being as follows: Figure 8 The terminal device or server shown. This embodiment uses this electronic device as an example for illustration. Figure 8 As shown, the electronic device includes: at least one processor 804; and a memory 802 communicatively connected to the at least one processor 804; wherein the memory 802 stores a computer program executable by the at least one processor 804, the computer program being executed by the at least one processor 804 to cause the at least one processor 804 to perform the steps in any of the above method embodiments. The electronic device may be located in at least one of a plurality of network devices in a computer network. The processor may be configured to execute the poster generation method via the computer program.

[0056] Alternatively, as those skilled in the art will understand, Figure 8 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 8 This does not limit the structure of the aforementioned electronic devices. For example, electronic devices may also include components that are more... Figure 8 The more or fewer components shown (such as network interfaces, etc.) or having the same Figure 8 The different configurations shown.

[0057] The memory 802 can be used to store software programs and modules, such as the program instructions / modules corresponding to the poster generation method and apparatus in this embodiment of the invention. The processor 804 executes various functional applications and data processing by running the software programs and modules stored in the memory 802, thereby realizing the aforementioned poster generation method. The memory 802 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 802 may further include memory remotely located relative to the processor 804, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. As an example, such as... Figure 8 As shown, the memory 802 may include, but is not limited to, the construction unit 702, retrieval unit 704, acquisition unit 706, and replacement unit 708 from the poster generation device described above. Furthermore, it may include, but is not limited to, other module units from the poster generation device described above, which will not be elaborated upon in this example.

[0058] Optionally, the transmission device 806 is configured to receive or send data via a network. Examples of the network include a wired network and a wireless network. In one example, the transmission device 806 includes a network interface controller (NIC) which can be connected to other network devices and routers through a network cable to communicate with the Internet or a local area network. In one example, the transmission device 806 is a radio frequency (RF) module which is configured to communicate with the Internet through a wireless manner. In addition, the electronic device further includes a display 808 and a connection bus 810 which is configured to connect various module components in the electronic device.

[0059] Obviously, those skilled in the art should understand that each unit or each step of the present application can be realized by a universal computing device, and they can be centralized on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be realized by program codes executable by a computing device, so that they can be stored in a storage device and executed by a computing device, or they can be respectively manufactured into each integrated circuit module, or multiple modules or steps among them can be manufactured into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

[0060] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of poster generation, characterized by, The method comprises the following steps: receiving a poster generation requirement and constructing a requirement structured vector based on the poster generation requirement; performing template retrieval based on the requirement structured vector to obtain a first poster template, and adjusting elements of the first poster template based on element adjustment instructions of a target user to obtain a second poster template; obtaining a plurality of target element contents corresponding to the second poster template; replacing element contents of the second poster template based on the plurality of target element contents to obtain a target poster.

2. The method of claim 1, wherein, Adjusting elements of the first poster template based on element adjustment instructions of a target user to obtain a second poster template comprises: obtaining first element information of the first poster template, wherein the first element information comprises element content, element area size, and area position information corresponding to the element area size; generating a first template element layer based on element layer generation instructions and the first element information; obtaining the element adjustment instructions and determining at least one element area to be adjusted in the first template element layer based on the element adjustment instructions; adjusting at least one of the element areas to be adjusted according to the element adjustment instructions to obtain the second poster template.

3. The method of claim 2, wherein, Generating a first template element layer based on element layer generation instructions and the first element information comprises: in the case where the element layer generation instructions indicate generating a floating template element layer, generating a reference template element layer based on the first element information, and obtaining a plurality of element areas in the reference template element layer; adaptively adding an identification area to each of the element areas to obtain a plurality of reference element areas; generating the first template element layer based on the plurality of reference element areas.

4. The method of claim 2, wherein, Replacing element contents of the second poster template based on a plurality of target element contents to obtain a target poster comprises: obtaining a plurality of second element areas in the second poster template, and pairing a plurality of target element contents with a plurality of second element areas respectively to obtain a plurality of content-area pairs, wherein each target element content corresponds to a second element area; inputting the plurality of content-area pairs into a content adjustment model respectively to output a plurality of replacement content-area pairs; replacing element contents in a plurality of second element areas according to the plurality of replacement content-area pairs to obtain the target poster.

5. The method of claim 4, wherein, Inputting the plurality of content-area pairs into a content adjustment model respectively to output a plurality of replacement content-area pairs comprises: determining any content-area pair as a current content-area pair, and obtaining area attribute information in the current content-area pair; performing semantic analysis on the content in the current content-area pair to obtain an analysis result; generating replacement content matching the content in the current content-area pair based on the analysis result and the area attribute information, wherein the semantic similarity between the replacement content and the content in the current content-area pair is greater than or equal to a preset threshold; replacing the content in the current content-area pair with the replacement content to obtain a replacement content-area pair, and determining content-area pairs other than the current content-area pair as current content-area pairs.

6. The method of claim 1, wherein, constructing a demand structured vector based on the poster generation demand, comprising: extracting keywords from the poster generation demand to obtain a plurality of poster keywords; performing semantic analysis on the poster generation demand to obtain a semantic analysis result; constructing demand structured data based on the plurality of poster keywords and the semantic analysis result; converting the demand structured data into the demand structured vector.

7. The method of claim 1, wherein, After obtaining the target poster, further comprising: constructing detection context information according to the target poster, wherein the detection context information comprises poster information and poster generation demand; inputting the target poster and the detection context information into a poster detection model to output a detection report and an optimization suggestion report.

8. The method of claim 7, wherein, inputting the target poster and the detection context information into a poster detection model to output a detection report and an optimization suggestion report, comprising: determining a plurality of detection types corresponding to the target poster based on the detection context information; performing poster detection on the target poster based on the detection context information and each detection type to obtain a detection result corresponding to each detection type and an optimization suggestion corresponding to each detection type; generating the detection report based on the detection result corresponding to each detection type, and generating the optimization suggestion report based on the optimization suggestion corresponding to each detection type.

9. A poster generation apparatus characterized by comprising: comprising: a construction unit configured to receive a poster generation demand and construct a demand structured vector based on the poster generation demand; a retrieval unit configured to perform template retrieval based on the demand structured vector to obtain a first poster template, and adjust elements of the first poster template based on element adjustment instructions of a target user to obtain a second poster template; an acquisition unit configured to acquire a plurality of target element contents corresponding to the second poster template; a replacement unit configured to replace element contents of the second poster template based on the plurality of target element contents to obtain a target poster.

10. An electronic device, comprising: The electronic device comprises at least one processor and a memory connected in communication with the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the poster generation method of any one of claims 1-8.

Images