Display device and interaction method based on work completion state

Abstract

The application discloses a display device and an interaction method based on a work completion state. The display device comprises a display, a user input interface and a controller. The user input interface collects a creation completion signal. The controller locks work information containing a work image after capturing the creation completion signal, analyzes work visible details and work completion degree information of the work information, and generates candidate evaluation text according to the analyzed information. Then, compliance verification containing modification guidance interception is performed on the text. If the verification is unqualified, the text is regenerated according to the verification result. If the verification is qualified, the text is pushed to the display as a target evaluation for display. The application distinguishes between a creation process state and a final draft completion state through a creation completion signal, generates an evaluation through real details of a picture, and intercepts modification type guidance language through a verification mechanism, so that optimization suggestions are avoided to be pushed to a completed work, the problem of hollow and template evaluation is solved, and the reliability of creation interaction feedback and the user creation experience are improved.

Description

Technical Field

[0001] This application relates to the field of display device technology, and more particularly to a display device and an interaction method based on the completion state of a work. Background Technology

[0002] Display devices are intelligent devices capable of presenting a user interface and supporting user interaction. Taking smart TVs as an example, smart TVs are television products based on Internet application technology, equipped with open operating systems and chips, possessing open application platforms, enabling two-way human-computer interaction, and integrating multiple functions such as audio-visual, entertainment, and data to meet diverse and personalized user needs.

[0003] Display devices can be equipped with AI creative assistants and digital painting platforms, integrating a multimodal painting assistance and interaction solution to provide real-time support for the user's creative process. This solution utilizes a multimodal image recognition model to analyze the canvas content and combines it with a generative model to output interactive feedback text. During the creation process, the system employs a continuous interaction mode, simultaneously providing guidance for modifications, supplementary drawing, and composition optimization, regardless of whether the user is in continuous editing or has completed their creation. The display device extracts the main information of the image through single-modal image recognition, directly inputs it into the generative model to generate feedback text, and displays this text via pop-up windows or voice prompts on the display device's interactive interface, while also storing it cached along with the artwork.

[0004] However, the above-mentioned drawing interaction feedback does not distinguish between the creative process and the completed work. As a result, even after the user confirms that the creation is complete, the system continues to output modification and optimization guidance content. Moreover, the feedback text is empty and lacks actual picture details to support it, thus affecting the reliability of the drawing interaction feedback. Summary of the Invention

[0005] This application provides a display device and an interaction method based on the completion state of a work to solve the problem of low reliability of session interaction feedback in display devices.

[0006] In a first aspect, this application provides a display device, comprising: monitor; The user input interface is configured to collect user input signals indicating completion of creation; The controller is electrically connected to the display and the user input interface, respectively, and the controller is configured to: The creation completion signal input by the user is captured through the user input interface; In response to the creation completion signal, the artwork information corresponding to the created artwork is locked, and the artwork information includes the artwork image; The artwork information is analyzed, and visible details and completion information are extracted from the artwork information; the visible details are the visual content identified from the artwork image, and the completion information is the judgment result representing the area that does not need to be drawn and the degree of content completeness; Candidate evaluation text is generated based on the visible details and completion information of the work. Perform compliance checks on the candidate evaluation texts, including at least checking for any disabling intent to guide users to modify the work. If the compliance verification result is unqualified, the candidate evaluation text is regenerated based on the verification result. If the compliance verification result is qualified, the candidate evaluation text is pushed to the display as the target evaluation text for display.

[0007] The above technical solution has the following beneficial effects or advantages: By using completion signals to distinguish between the ongoing and finished stages of creation, the system avoids outputting modification and optimization guidance content after the user confirms the final draft, solving the problems of not being able to distinguish the creation stages and the inconvenience of final feedback. Evaluation content is generated by extracting realistic visual details from the artwork image and using objective completeness assessment results, reducing empty and templated evaluations at the source. Simultaneously, a compliance verification mechanism including disabling intent interception and an automatic retry loop for unqualified evaluations improves the compliance and authenticity of the output evaluations, enhancing the user experience.

[0008] In some embodiments of this application, the controller is specifically configured to execute the locked artwork information corresponding to the created artwork as follows: In response to the creation completion signal, freeze a snapshot of the created work; The work image, work summary, and creation process evidence are extracted from the work snapshot as the work information; the work summary is a structured text description of the work content, and the creation process evidence is an operation record of the entire creation process.

[0009] The above technical solution has the following beneficial effects or advantages: Upon detecting a completion signal, the current snapshot of the work is immediately frozen, fixing all materials in the final draft state to prevent subsequent editing operations from damaging the baseline data for evaluation and analysis. At the same time, the work image, structured summary, and complete creation operation record are used as the basis for analysis, integrating multi-dimensional creation evidence to provide a data foundation for subsequent multimodal semantic recognition, thereby improving the accuracy of image detail extraction and work integrity determination, so that the generated evaluation can completely restore the user's entire creation process and image content.

[0010] In some embodiments of this application, the controller is specifically configured to perform the extraction of visible details and completion information of the artwork from the artwork information as follows: The artwork image, artwork summary, and creative process evidence are jointly identified using a multimodal semantic model, and the main subject, color features, background elements, composition layout, and brushstroke details are extracted as visible details of the artwork. The completion information of the artwork is determined based on whether there are areas to be added to the artwork image and the degree of completeness of the content.

[0011] The above technical solution has the following beneficial effects or advantages: By using a multimodal semantic model to jointly identify three types of data—images, text summaries, and operation records—compared to relying solely on single-image recognition, this approach can accurately extract multi-level visual details of the image by combining the user's entire creative process. At the same time, it can objectively determine the completeness of the work based on the blank areas to be drawn on the image, providing a basis for distinguishing between completed and modified evaluations, thereby enhancing the relevance of completed evaluations.

[0012] In some embodiments of this application, the controller is specifically configured to generate candidate evaluation text based on the visible details of the work and the work's completion information as follows: Based on the main subject, color features, and background elements in the visible details of the artwork, generate observation and evaluation statements; Based on the brushstroke details and composition of the artwork, generate sincere words of encouragement. The observation and evaluation statements are combined with the sincere encouragement statements to form the candidate evaluation text.

[0013] The above technical solution has the following beneficial effects or advantages: The evaluation text is divided into two parts: objective observation statements and encouraging statements on creative efforts. The corresponding content is generated by the surface visual elements and the deep creative brushstroke layout of the work, respectively. This ensures that the evaluation content comes from the real details of the work and improves the user's interactive experience after completing the creation.

[0014] In some embodiments of this application, the controller performs compliance verification, specifically configured as follows: Verify whether the candidate evaluation text contains no less than a preset number of real, visible details to obtain a first verification result; The second verification result is obtained by verifying whether the candidate evaluation text contains encouraging statements that fit the content of the work and whether there are any templated phrases. The third verification result is obtained by verifying whether there are any statements in the candidate evaluation text that contain the intention to guide users to redraw, adjust or optimize their works. The compliance verification result is determined based on the first verification result, the second verification result, and the third verification result.

[0015] The above technical solution has the following beneficial effects or advantages: A three-layer serial verification logic is set up to check the richness of details, the relevance of encouraging words, and the modification of the guidance's disabling intent. Figure 3 The evaluation text quality is comprehensively constrained across multiple dimensions. The evaluation is judged to be compliant by combining the results of multi-dimensional verification. It can filter out unqualified evaluations that are insufficient in detail, overly templated, or include modification suggestions, and unify the output standard of the completed evaluation.

[0016] In some embodiments of this application, the controller is specifically configured to perform the regeneration of the candidate evaluation text based on the verification result as follows: When the compliance verification result is unqualified, the dimension information that failed the verification is summarized to generate rejection reason information; The rejection reason information and the rejected candidate evaluation text are input into the evaluation generation model to regenerate the updated candidate evaluation text. The compliance check is performed again on the updated candidate evaluation text.

[0017] The above technical solution has the following beneficial effects or advantages: When the evaluation fails the verification, the non-compliant items are summarized to generate clear reasons for rejection and are simultaneously input into the evaluation generation model. This provides the model with corrective constraints, prevents the model from repeatedly generating evaluation texts with similar defects, and automatically performs a second verification on the updated evaluation, thereby continuously improving the compliance rate and quality of the output evaluation.

[0018] In some embodiments of this application, the controller is further configured to: The target evaluation text is converted into speech reading data; The display simultaneously presents the text content of the target evaluation text and the playback status of the voice reading data.

[0019] The above technical solution has the following beneficial effects or advantages: Converting the reviewed target evaluation text into voice reading data can expand the voice interaction feedback channels, adapt to users who have difficulty reading text, such as children and the elderly, and simultaneously display the text content and voice playback status on the display screen, realizing the simultaneous output of evaluation through both text and voice channels, and enriching the interactive forms after the creation is completed.

[0020] In some embodiments of this application, the controller is further configured to: The target evaluation text is bound to the corresponding creative work to generate a work version record; the work version record includes evaluation content, review items, and verification results; The version records of the works are stored in the works version library.

[0021] The above technical solution has the following beneficial effects or advantages: By binding compliance evaluations and complete review and verification information with corresponding creative works to generate independent version records and archiving them, the full lifecycle management of creative works can be realized, making it convenient for users to retrieve and review historical works, along with accompanying evaluations and review records at any time.

[0022] In some embodiments of this application, the controller executes the creation completion signal captured through the user input interface, specifically configured as follows: Detect at least one of the following types of signals as the creation completion signal: Touch signals from users clicking the "Done" button, voice signals from users completing input, save status signals triggered by saving the work, and static status signals from prolonged periods without editing.

[0023] The above technical solution has the following beneficial effects or advantages: It is compatible with four different types of creation completion signals: touch click, voice, save action, and long-term static display. It covers the final draft recognition scenarios where users actively confirm and the system passively judges. It is compatible with various human-computer interaction hardware such as touch screens, microphones, and remote controls, and can match the creation operation habits of different users to improve the interaction adaptability and ease of use.

[0024] Secondly, this application also provides an interaction method based on the completed state of a work, including: Capture the creation completion signal input by the user through the user input interface; In response to the creation completion signal, the artwork information corresponding to the created artwork is locked, and the artwork information includes the artwork image; The artwork information is analyzed, and visible details and completion information are extracted from the artwork information; the visible details are the visual content identified from the artwork image, and the completion information is the judgment result representing the area that does not need to be drawn and the degree of content completeness; Candidate evaluation text is generated based on the visible details and completion information of the work. Perform compliance checks on the candidate evaluation texts, including at least checking for any disabling intent to guide users to modify the work. If the compliance verification result is unqualified, the candidate evaluation text is regenerated based on the verification result. If the compliance verification result is qualified, the candidate evaluation text will be displayed as the target evaluation text.

[0025] The above technical solution has the following beneficial effects or advantages: This interaction method, based on the completed artwork state, avoids continuously providing modification and optimization guidance content after the artwork is finalized. It extracts realistic visual details from the artwork image and generates evaluation content based on objective completeness assessment results, fundamentally reducing empty and templated evaluations. Simultaneously, a compliance verification mechanism enhances the compliance and authenticity of the output evaluations, thereby improving the user experience. Attached Figure Description

[0026] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram illustrating the operation scenarios of a display device provided in some embodiments of this application; Figure 2 This is a schematic diagram of the hardware configuration of a display device provided in some embodiments of this application; Figure 3 This is a schematic diagram of the software configuration of a display device provided in some embodiments of this application; Figure 4 A flowchart illustrating an interaction method based on the completion state of a work, provided for some embodiments of this application; Figure 5 These are schematic diagrams illustrating the display effects of prompt information provided in some embodiments of this application; Figure 6 A flowchart illustrating the compliance verification process provided for some embodiments of this application; Figure 7 A flowchart illustrating the process of determining verification results provided in some embodiments of this application; Figure 8 A schematic diagram illustrating the speech reading process provided for some embodiments of this application; Figure 9 These are schematic diagrams illustrating the display and playback effects of target evaluation text provided in some embodiments of this application. Figure 10 A schematic diagram illustrating the process of generating a work version record for some embodiments of this application; Figure 11 This is an interactive diagram illustrating an interactive method based on the completion state of a work, provided in some embodiments of this application. Detailed Implementation

[0028] The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with this application. They are merely examples of systems and methods consistent with some aspects of this application as detailed in the claims.

[0029] In this embodiment, display device 200 generally refers to a device with screen display and data processing capabilities. For example, display device 200 includes, but is not limited to, smart TVs, mobile terminals, computers, monitors, advertising screens, wearable devices, virtual reality devices, augmented reality devices, etc.

[0030] Figure 1 This is a schematic diagram illustrating an operational scenario between a display device and a control device provided in some embodiments of this application. For example... Figure 1 As shown, users can operate the display device 200 via touch operation, mobile terminal 300, and control device 100. For example, control device 100 can be a remote control, stylus, gamepad, etc.

[0031] The mobile terminal 300 can function as a control device for human-computer interaction between the user and the display device 200. It can also function as a communication device for establishing a communication connection with the display device 200 and exchanging data. In some embodiments, the mobile terminal 300 can have software applications installed on it and communicate with the display device 200 via network communication protocols to achieve one-to-one control and data communication. Furthermore, it can transmit audio and video content displayed on the mobile terminal 300 to the display device 200 for synchronized display.

[0032] like Figure 1 The diagram also shows that the display device 200 communicates with the server 400 via various communication methods. This allows the display device 200 to communicate via a local area network (LAN), a wireless local area network (WLAN), and other networks.

[0033] Display device 200 can provide broadcast television reception function, and can also be equipped with intelligent network television function that provides computer support function, including but not limited to network television, smart television, Internet Protocol television (IPTV), etc.

[0034] Figure 2 Provided for some embodiments of this application Figure 1 Hardware configuration block diagram of display device 200.

[0035] In some embodiments, the display device 200 may include at least one of a tuner 210, a communication device 220, a detector 230, a device interface 240, a controller 250, a display 260, an audio output device 270, a memory, a power supply, and a user input interface 280.

[0036] In some embodiments, detector 230 is used to collect signals from the external environment or signals interacting with the outside world. For example, detector 230 may include millimeter-wave radar, which can be used to detect whether a user is present within a preset range. Detector 230 may also include a voice acquisition unit to collect voice commands input by the user.

[0037] In some embodiments, the display 260 includes display function components for presenting images and driving components for driving image display. The display 260 is used to receive and display image signals output from the controller 250. For example, the display 260 can be used to display video content, image content, menu control interface components, and user control UI interfaces, etc.

[0038] In some embodiments, the communication device 220 is a component used to communicate with external devices or the server 400 according to various communication protocol types. The display device 200 may have multiple communication devices 220 depending on the supported communication methods. For example, when the display device 200 supports wireless network communication, it may have a communication device 220 with WiFi functionality. When the display device 200 supports Bluetooth connectivity, it needs to have a communication device 220 with Bluetooth functionality.

[0039] The communication device 220 enables the display device 200 to communicate with external devices or the server 400 via wireless or wired connections. Wired connections utilize data cables, interfaces, or other components to connect the display device 200 to external devices. Wireless connections utilize wireless signals or wireless networks. The display device 200 can directly establish a connection with external devices or indirectly through gateways, routers, or other connection devices.

[0040] In some embodiments, the controller 250 may include at least one of a central processing unit, a video processor, an audio processor, a graphics processor, and a power processor, and a first to an nth interface for input / output. The controller 250 controls the operation of the display device and responds to user operations through various software control programs stored in memory. The controller 250 controls the overall operation of the display device 200.

[0041] In some embodiments, the controller 250 and the tuner 210 may be located in different separate devices, that is, the tuner 210 may also be located in an external device of the main device where the controller 250 is located, such as an external set-top box.

[0042] In some embodiments, a user can input user commands through a graphical user interface (GUI) displayed on a display 260, and the user input interface receives user input commands through the graphical user interface (GUI).

[0043] In some embodiments, the audio output device 270 can be a built-in speaker of the display device 200 or an external audio output device connected to the display device 200. For the external audio output device connected to the display device 200, the display device 200 may also be provided with an external audio output terminal, through which the audio output device can be connected to the display device 200 to output sound from the display device 200.

[0044] In some embodiments, the user input interface 280 can be used to receive instructions input by a user. The user input interface 280 may include at least one of a microphone, touchpad, sensor, remote control, etc. The display device 200 can then receive user-input instructions based on the user input interface 280 to perform interactive functions with the user.

[0045] In some embodiments, to enable user interaction, the display device 200 may run an operating system. An operating system is a computer program that manages and controls the hardware and software resources of the display device 200. The operating system can control the display device to provide a user interface; for example, the operating system can directly control the display device to provide a user interface, or it can provide a user interface by running applications. The operating system also allows users to interact with the display device 200.

[0046] It should be noted that the operating system can be a native operating system based on a specific operating platform, a third-party operating system that is deeply customized based on a specific operating platform, or an independent operating system specifically developed for display devices.

[0047] An operating system can be divided into different modules or levels based on the functions it implements, for example... Figure 3 As shown, in some embodiments, the system is divided into four layers, from top to bottom: the Applications layer (referred to as the "Application Layer"), the Application Framework layer (referred to as the "Framework Layer"), the System Library layer, and the Kernel layer.

[0048] In some embodiments, the application layer provides services and interfaces for applications, enabling the display device 200 to run applications and interact with the user based on the applications. The application layer may contain at least one application, which may be a built-in Windows program, system settings program, or clock program of the operating system; or it may be an application developed by a third-party developer. In specific implementations, the application packages in the application layer are not limited to the examples above.

[0049] The framework layer provides application programming interfaces (APIs) and a programming framework for applications. The application framework layer includes predefined functions. It acts as a central processing unit, determining the actions taken by applications within the application layer. Through the API, applications can access system resources and obtain system services during execution.

[0050] like Figure 3 As shown, the application framework layer in this embodiment includes a view system, managers, and content providers. The view system designs and implements the application's interface and interactions, and includes lists, grids, text boxes, and buttons. The managers include at least one of the following modules: an activity manager for interacting with all running activities in the system; a location manager for providing system services or applications with access to system location services; a package manager for retrieving various information related to application packages currently installed on the device; a notification manager for controlling the display and clearing of notification messages; and a window manager for managing icons, windows, toolbars, wallpapers, and desktop widgets on the user interface.

[0051] In some embodiments, the Activity Manager manages the lifecycle of individual applications and common navigation and back functions, such as controlling application exit, opening, and back actions. The Window Manager manages all window programs, such as obtaining the screen size, determining if a status bar is present, locking the screen, capturing the screen, and controlling changes to the display window, such as shrinking the display window, shaking the display, or distorting the display.

[0052] In some embodiments, the system runtime library layer can provide support for the framework layer. When the framework layer is used, the operating system runs the instruction library contained in the system runtime library layer, such as the C / C++ instruction library, to implement the functions to be performed by the framework layer.

[0053] In some embodiments, the kernel layer is a functional layer situated between the hardware and software of the display device 200. The kernel layer can implement functions such as hardware abstraction, multitasking, and memory management. For example, ... Figure 3 As shown, hardware drivers can be configured in the kernel layer. The kernel layer can contain at least one of the following drivers: audio driver, display driver, Bluetooth driver, camera driver, WIFI driver, USB driver, HDMI driver, sensor driver (such as fingerprint sensor, temperature sensor, pressure sensor, etc.), and power driver, etc.

[0054] In some embodiments, the display device 200 may be equipped with an AI creation assistant and a digital painting platform, which integrates a multimodal painting assistance interaction scheme to provide real-time assistance to the user's creative process. The painting assistance interaction scheme uses a multimodal image recognition model to analyze the canvas content and combines it with a generative large model to output interactive feedback text.

[0055] During the creation process, the system adopts a continuous interactive mode. Regardless of whether the user is in continuous editing or has completed the creation process, the system synchronously outputs inspirational content such as guidance for modification, supplementary drawing, and composition optimization. After extracting the main information of the image through image single-modal recognition, the display device 200 directly inputs the generated model to produce feedback text, which is then displayed through a pop-up window or pushed to the interactive interface of the display device 200 via voice, and simultaneously cached and stored along with the artwork.

[0056] However, the aforementioned drawing interaction feedback did not distinguish between the in-process and completed stages of the artwork. As a result, even after the user confirmed the completion of their work, the system continued to output guidance content for modification and optimization. Furthermore, the feedback text was vague and lacked support from actual visual details. For example, it only outputted generic template phrases such as "Great drawing!" and "The colors are beautiful," failing to make the user feel that the details of their artwork were being recognized, thus affecting the reliability of the drawing interaction feedback.

[0057] To address the aforementioned issues, this application provides a display device and an interaction method based on the completion state of a work. By recognizing the completion state, extracting details from the real-world image, and implementing a compliance verification mechanism, this method improves upon the problems of pushback suggestions for modification and empty, template-based evaluations in the feedback of the completion state.

[0058] In some embodiments, the display device 200 includes at least a display 260, a user input interface 280, and a controller 250. The display 260 is used to display a user interface; the user input interface 280 is configured to acquire a creation completion signal input by the user; and the controller 250 is electrically connected to the display 260 and the user input interface 280, respectively.

[0059] In some embodiments, the user input interface 280 may include at least one input component such as a touch screen, microphone, remote control receiver, mouse, or keyboard.

[0060] Figure 4 This is a flowchart illustrating an interaction method based on the completion state of a work, provided for some embodiments of this application. For example... Figure 4 As shown, controller 250 is configured to perform the following steps: S401. Capture the creation completion signal input by the user through the user input interface.

[0061] The controller 250 detects the creation completion signal input by the user through the user input interface 280.

[0062] In some embodiments, the creation completion signal includes at least one of the following types of signals: a touch signal when the user clicks the completion button, a voice signal when the user completes input, a save status signal triggered by saving the work, and a static status signal when there is no editing operation for a long time.

[0063] For example, after a user completes their work on a digital painting platform, they can input a touch signal by clicking the "Finish" button on the interface; or input a voice signal by saying "I'm done drawing"; or trigger a save status signal when saving the work; or automatically generate a static status signal when the system detects that the user has not performed any editing operations for more than a preset time (such as 5 minutes).

[0064] In some implementations, when the controller 250 detects multiple creation completion signals, it processes the signals according to a preset priority. Touch signals and voice signals have higher priority than save status signals and idle status signals because touch signals and voice signals typically represent a clear intention from the user to actively confirm the completion of the creation, while save status signals and idle status signals may be automatically triggered by the system, potentially leading to misjudgment. When the controller 250 detects multiple signals within the same time window, it prioritizes the higher-priority signal and discards the others to avoid repeatedly triggering the artwork information locking process.

[0065] In addition, the controller 250 verifies the validity of the captured creation completion signals. For touch signals, it verifies whether the corresponding touch area is the preset completion button area and whether the touch duration exceeds a minimum duration threshold (e.g., 200 milliseconds). For voice signals, after converting the input speech recognition model into text, it matches it with a preset completion instruction semantic template. Only when the semantic similarity exceeds a preset threshold is it considered a valid creation completion signal. For save status signals, it verifies whether the save operation corresponding to the save status signal is a substantial update to the work content (e.g., the number of drawing operations exceeds a minimum threshold), rather than an empty save with no content. For static status signals, it verifies whether the static duration exceeds a preset no-operation threshold (e.g., 5 minutes) and whether there is any drawn content on the canvas. Through the above validity verification, it can effectively filter out false trigger signals caused by accidental touches, environmental noise, system anomalies, etc., and improve the accuracy and robustness of creation completion signal detection.

[0066] This embodiment of the disclosure detects multiple types of creation completion signals, covering two finalization recognition scenarios: user active confirmation and system passive judgment, adapting to the creation operation habits of different users.

[0067] S402. In response to the completion signal of creation, lock the work information corresponding to the created work.

[0068] After capturing the creation completion signal, the controller 250 responds to the creation completion signal by locking the artwork information corresponding to the currently created artwork, which includes the artwork image.

[0069] In some embodiments, in response to a creation completion signal, the controller 250 freezes a snapshot of the created work and extracts the work image, work summary, and evidence of the creation process from the snapshot as work information. The work summary is a structured text description of the work's content, and the evidence of the creation process is a record of the entire creation process.

[0070] For example, the artwork image is a screenshot or rendered image of the canvas when the creation is completed; the artwork summary includes information such as the artwork title, creation time, and type of tools used; evidence of the creation process includes operation records such as the coordinates, color, brush parameters, and timestamps of each drawing operation.

[0071] In some implementations, the process of freezing the artwork snapshot includes: upon detecting a completion signal, the controller 250 pauses the canvas rendering thread, writes the image data in the current frame buffer to a read-only memory area, and generates the artwork image. Simultaneously, the controller 250 iterates through the complete operation log of the current creation session, serializing and storing the timing records of all drawing operations as evidence of the creation process. Each operation record includes information such as operation type (e.g., pen stroke, movement, pen lift, fill, erase), operation coordinates, color parameters, brush type and size, and timestamp. The artwork summary is automatically generated by the controller 250 based on the operation log, including the artwork title (if the user has named it during the creation process), total creation time, total number of drawing operations, a list of brush types used, a list of colors used in the palette, and information such as canvas size and resolution.

[0072] By encapsulating the above three types of information into a unified data structure, it is ensured that subsequent multimodal semantic models can read and parse them using a standardized interface. Furthermore, after freezing the snapshot, the controller 250 marks the editing state of the artwork as read-only and locked. If the user attempts to continue editing the canvas during the evaluation generation process, a prompt message will be displayed on the monitor 260, such as... Figure 5 The displayed message informs the user that the evaluation is being generated and that they can continue operating only after the evaluation is completed, thus ensuring the consistency and integrity of the evaluation analysis benchmark data. When the evaluation generation process ends (regardless of whether the verification passes or fails), the controller 250 automatically unlocks the editing mode and restores the normal editing function of the canvas.

[0073] In this embodiment, freezing a snapshot of the work fixes all materials in the final draft state, preventing subsequent editing operations from damaging the benchmark data for evaluation and analysis.

[0074] S403. Analyze the artwork information and extract the visible details and completion information of the artwork from the artwork information.

[0075] The controller 250 parses the locked artwork information and extracts visible details and artwork completion information. Visible details are the visual content identified from the artwork image, and artwork completion information is the judgment result representing the areas that do not need to be drawn and the completeness of the content.

[0076] In some embodiments, the controller 250 uses a multimodal semantic model to jointly identify the artwork image, artwork summary and evidence of the creation process, and extracts the main subject, color features, background elements, composition layout and brushstroke details as visible details of the artwork.

[0077] In some embodiments, the controller 250 determines the completion information of the artwork based on whether there are areas to be drawn in the artwork image and the degree of content completeness.

[0078] For example, a multimodal semantic model can use the CLIP (Contrastive Language-Image Pre-training) model or the BLIP (Bootstrapping Language-Image Pre-training) model to perform cross-modal alignment between the artwork image and the artwork summary text, and combine the operation sequence in the evidence of the creation process to jointly identify multi-level visual details in the image.

[0079] The main subject of the image may include people, animals, plants, buildings, and other major objects; color characteristics may include the main color tone, color matching, and contrast; background elements may include background content such as the sky, grass, and indoor scenes; composition layout may include composition information such as the position of the main subject, the proportion of the image, and visual hierarchy; brushstroke details may include brushstroke information such as line thickness, brush type, and drawing pressure.

[0080] Regarding the completion information of the work, the controller 250 uses an image segmentation algorithm to detect whether there are large blank areas or incomplete sketch lines in the work image. If they exist, the work is determined to be incomplete; if they do not exist, the work is determined to be complete.

[0081] In some implementations, the multimodal semantic model preprocesses the data for each modality before receiving the artwork image, artwork summary, and evidence of the creative process. For the artwork image, the controller 250 scales it to a uniform resolution (e.g., 224 x 224 pixels) and normalizes the pixel values, mapping them to the 0-1 range to suit the model's input requirements. For the artwork summary, the controller 250 converts the structured fields into natural language descriptive text and encodes it into a text feature vector using a pre-trained language model. For the evidence of the creative process, the controller 250 extracts temporal features and brushstroke trajectory features from the operation sequence, converting discrete operation records into continuous temporal feature vector representations.

[0082] Subsequently, the multimodal semantic model fuses feature vectors from the three modalities through a cross-modal attention mechanism, enhancing image features, text features, and operation trajectory features to improve the accuracy of image detail recognition. For example, when the artwork summary records that the user used a watercolor brush to paint flowers, the text information can guide the model to assign higher attention weights to flower-related brushstroke patterns and color features in the image features, thereby more accurately extracting the main features and brushstroke details of the flowers. Simultaneously, the brushstroke movement trajectories in the evidence of the creative process can help the model distinguish between foreground subjects and background elements in the image, because foreground subjects usually correspond to more drawing operations and denser brushstroke trajectories, while background elements often involve fewer drawing operations or use large-area filling methods.

[0083] To determine the completion status of a work, the controller 250 employs a semantic segmentation-based blank area detection algorithm. In some embodiments, the controller 250 inputs the work image into a pre-trained semantic segmentation model, divides the image into multiple semantic regions, and calculates the pixel coverage of each region. If the pixel coverage of a certain semantic region is lower than a preset threshold (e.g., 10%), and this region is located around the main subject of the image rather than in the natural white space at the edge of the canvas, then this region is marked as a region to be supplemented. When there are no regions to be supplemented in the work image, the controller 250 determines that the work completion status is complete; conversely, if there is at least one region to be supplemented, it is determined to be incomplete, and the position coordinates and area percentage of the region to be supplemented are recorded in the completion status information to provide data for subsequent evaluation generation.

[0084] S404. Generate candidate evaluation text based on the visible details and completion information of the work.

[0085] The controller 250 generates candidate evaluation text based on the extracted visible details and completion information of the work.

[0086] In some embodiments, the controller 250 generates observation and evaluation statements based on the main subject, color features, and background elements in the visible details of the artwork; generates sincere encouragement statements based on the brushstroke details and composition layout in the visible details of the artwork; and combines the observation and evaluation statements with the sincere encouragement statements into candidate evaluation text.

[0087] For example, the observation and evaluation statement could be "I see that you drew a little girl in a red dress standing next to a green tree with a blue sky in the background", where "little girl in red dress" comes from the subject recognition result, "green tree" comes from the background element recognition result, and "blue sky" comes from the color feature recognition result.

[0088] A sincere encouraging statement could be, "You used very detailed lines to depict the little girl's hair, and the composition of the picture places the little girl in the middle, creating a sense of depth." Here, "detailed lines" comes from the brushstroke detail recognition results, while "standing in the middle" and "sense of depth" come from the composition layout recognition results.

[0089] In some implementations, when generating candidate evaluation text, the controller 250 first constructs an evaluation generation prompt template, filling in the corresponding fields of the template with various features from the visible details of the artwork. The evaluation generation prompt template includes observation dimension fields (main subject, color features, background elements) and encouragement dimension fields (brushstroke details, composition layout), with corresponding feature description text appended to each field. The controller 250 inputs the filled prompt template into the evaluation generation model, which then generates observation evaluation statements and sincere encouragement statements, respectively.

[0090] In some embodiments, during the sentence assembly stage, the controller 250 combines the evaluation statements first, followed by the encouraging statements, and inserts natural transition words between the two sentences to make the candidate evaluation text read smoothly and coherently. Furthermore, the controller 250 adapts the language style of the evaluation text based on the creator's user profile information. User profile information can be obtained from user account registration information or historical creative behavior data, including age group, historical creative types, and creative proficiency.

[0091] For example, when the creator is a child, the evaluation text uses simple and easy-to-understand short sentences and lively interjections, avoiding the use of overly professional art terms; when the creator is an adult user or has a high level of creative proficiency, the evaluation text uses a more formal expression and appropriately introduces professional art terms (such as color contrast, golden ratio composition, etc.) to enhance the professionalism and relevance of the evaluation content.

[0092] In this embodiment of the disclosure, by splitting the evaluation text into objective visual observation statements and statements encouraging creative efforts, it is ensured that all evaluation content comes from the true details of the work.

[0093] In some implementations, the evaluation generation model can employ a generative pre-trained language model based on the Transformer architecture, fine-tuned on a corpus in the field of art education. This fine-tuned corpus includes a large number of artwork evaluation texts written by professional art teachers, each labeled with corresponding visual feature tags (such as subject type, color style, composition, etc.). Through domain-specific fine-tuning, the evaluation generation model can generate more artistically professional and educationally inspiring evaluation statements, avoiding the output of generalized evaluations unrelated to the artwork's content. Furthermore, the controller 250 incorporates a temperature parameter to control the diversity of generated evaluation texts. While ensuring the accuracy of the evaluation content, this allows for the generation of different evaluation texts for the same artwork in multiple evaluation processes, preventing users from receiving identical evaluations after multiple creations and reducing the novelty of the interactive experience.

[0094] S405. Perform compliance verification on candidate evaluation texts.

[0095] The controller 250 performs compliance checks on the generated candidate evaluation text, including at least checking for any disabled intents that guide users to modify their work.

[0096] Figure 6 This is a flowchart illustrating the compliance verification process provided for some implementations of this application. For example... Figure 6As shown, in some embodiments, the compliance verification performed by the controller 250 includes three layers of verification: First, verifying whether the candidate evaluation text contains at least a preset number (e.g., 2) of genuinely visible details, obtaining a first verification result (S601); second, verifying whether the candidate evaluation text contains encouraging statements that fit the work content and does not contain templated language, obtaining a second verification result (S602); third, verifying whether the candidate evaluation text contains statements that prohibit the user from redrawing, adjusting, or optimizing the work, obtaining a third verification result (S603). The controller 250 determines the final verification result based on the first verification result, the second verification result, and the third verification result (S604).

[0097] For example, templated phrases include general comments such as "Great drawing," "Beautiful colors," and "Clear picture," which do not refer to specific details of the picture; prohibited intention statements include statements such as "You can draw some more flowers" and "I suggest adjusting the colors," which guide users to continue modifying their work.

[0098] In some implementations, the controller 250 also supports dynamic adjustment of threshold parameters for each layer of compliance verification. The controller 250 statistically analyzes the pass rate and false positive rate of each layer of verification in historical evaluation data. When the false positive rate of a certain layer of verification exceeds a preset upper limit, the similarity threshold for that layer of verification is automatically adjusted. For example, if the false positive rate of templated phrase comparison in the second layer of verification is high—that is, if non-template-based personalized encouragement statements are misjudged as templated phrases—the semantic similarity threshold is appropriately reduced (e.g., from 0.85 to 0.80) to reduce false positives.

[0099] In this embodiment of the disclosure, a dynamic threshold adjustment mechanism is used to enable the compliance verification criteria to adapt to changes in the capabilities of the evaluation generation model and the diversity of user-created content, thereby continuously maintaining the accuracy and rationality of the verification results.

[0100] Figure 7 This is a flowchart illustrating the process of determining verification results provided in some embodiments of this application. For example... Figure 7 As shown, in some embodiments, the controller 250 determines the compliance verification result based on the first verification result, the second verification result, and the third verification result. When any one of the three verification layers fails, the verification result is determined to be unqualified.

[0101] In some implementations, during the first-level verification, the controller 250 matches candidate evaluation text with visible details of the work, counting the number of actual visible details referenced in the evaluation text. Matching methods include keyword matching and semantic similarity calculation. Keyword matching compares entity nouns in the visible details of the work with the evaluation text to detect identical or synonymous words. Semantic similarity calculation encodes each statement in the evaluation text and the description text of each visible detail into a semantic vector, calculates the cosine similarity between the vectors, and determines that the statement references the visible detail when the similarity exceeds a preset threshold (e.g., 0.75). If the number of references is less than a preset number (e.g., 2), the first verification result is a failure.

[0102] In the second layer of verification, controller 250 maintains a templated phrase database, which stores general evaluation templates and their corresponding semantic vector representations. Controller 250 performs a semantic comparison between the encouraging statements in the candidate evaluation text and the templated phrase database. If the similarity exceeds a preset threshold (e.g., 0.85), the encouraging statement is determined to be templated, and the second verification result is "failed." Simultaneously, controller 250 also checks whether the encouraging statement has a semantic connection with the brushstroke details or composition layout visible in the artwork. If there is no connection, the encouraging statement is determined not to fit the content of the artwork.

[0103] In the third layer of verification, controller 250 maintains a prohibited intent keyword library and an intent classification model. The prohibited intent keyword library includes words and their synonyms that guide users to continue creating, such as "redraw," "supplement," "adjust," "optimize," "modify," "add," "suggest," "try," and "can be improved." The intent classification model classifies the overall semantics of candidate evaluation texts based on a pre-trained language model to determine if the text contains an intent to guide users to modify their work. Controller 250 first detects whether the candidate evaluation text contains prohibited intent keywords through keyword matching. If a match is found, the third verification result is directly determined to be a failure. If no keyword match is found, the candidate evaluation text is input into the intent classification model for deep semantic analysis. When the probability of prohibited intent output by the model exceeds a preset threshold (e.g., 0.6), the third verification result is also determined to be a failure. Through the dual detection mechanism of keyword matching and intent classification model, explicit and implicit modification guidance intents can be effectively identified.

[0104] S406. If the compliance verification result is unqualified, then the candidate evaluation text shall be regenerated based on the verification result.

[0105] When the compliance check result is unqualified (i.e., any layer of check fails), controller 250 summarizes the dimension information that failed the check and generates rejection reason information. The rejection reason information, along with the rejected candidate evaluation text, is input into the evaluation generation model to regenerate updated candidate evaluation text. Compliance check is then performed again on the updated candidate evaluation text, and this process is repeated until the check passes or the preset maximum number of retries is reached.

[0106] For example, if the first verification result is a failure (insufficient number of real visible details), the rejection reason information is "the evaluation text does not contain enough real image details"; if the third verification result is a failure (the existence of statements with the intention to disable), the rejection reason information is "the evaluation text contains statements that guide the user to modify the work".

[0107] In some implementations, the controller 250 sets a maximum regeneration threshold (e.g., 3 times). When the regeneration count reaches the threshold and the compliance check is still not passed, the controller 250 selects the evaluation text that best matches the visible details of the current work from a preset high-quality evaluation template library as the target evaluation text. The evaluation texts in the high-quality evaluation template library have been pre-verified for compliance and labeled with applicable work feature tags. The controller 250 selects the best template by calculating the matching degree between the visible details of the current work and each template feature tag.

[0108] In other implementations, if the compliance check still fails after the number of regeneration attempts reaches a threshold, the controller 250 extracts and combines the validated portions of the last generated candidate evaluation text into the target evaluation text, retaining only the validated sentence fragments and discarding the uncaught parts. By setting a maximum number of retries and a fallback strategy, the evaluation generation process is prevented from entering an infinite loop, while ensuring that evaluation text meeting basic quality requirements can still be output even in extreme cases.

[0109] Furthermore, in some embodiments, the controller 250 records the rejection reasons and generation parameter change information each time it is regenerated, forming a review tracking log. The review tracking log includes information such as the content of the candidate evaluation text generated each time, the verification dimensions that failed, the rejection reason information, and the content of the prompt words adjusted in the model input, which facilitates the subsequent analysis and optimization of the performance of the evaluation generation model, and can also be used to track the entire evaluation generation process of a specific work.

[0110] S407. If the compliance verification result is qualified, the candidate evaluation text will be pushed to the display as the target evaluation text for display.

[0111] Once the compliance verification is passed, the controller 250 pushes the candidate evaluation text as the target evaluation text to the user interface of the display 260 for display.

[0112] In some embodiments, the controller 250 may display the target evaluation text on the display 260 in the form of a pop-up card, and may also enhance the display experience with animation effects.

[0113] In some implementations, the pop-up card is overlaid on the current canvas with a semi-transparent overlay. The card contains the full content of the target evaluation text, formatted with a readable font size and line spacing. A confirmation button is located at the bottom of the card; when the user clicks it, the card fades out with a fade-out animation, the overlay is removed simultaneously, and the canvas returns to an editable state.

[0114] In some implementations, the card automatically closes if the user does not click the button within a preset time (e.g., 10 seconds). Simultaneously, the controller 250 freezes the canvas's editing functions during card display to prevent accidental drawing operations while the user is reading the review. When the target review text is too long and exceeds the card's display area, the card supports vertical scrolling, and a scrolling progress indicator is displayed on the right side of the card to help the user track their reading progress.

[0115] In this embodiment, the creation process is distinguished from the completion state by the creation completion signal. In the completion state, an evaluation is generated based on the details of the real picture and the modification-type guiding words are intercepted through compliance verification. This avoids pushing optimization suggestions even after the work is completed. At the same time, it solves the problem of empty and templated evaluations and improves the reliability of creative interaction feedback.

[0116] In some embodiments, when the display device 200 establishes a network connection with the server 400 through the communication device 220, the controller 250 can upload the work image, work summary and evidence of the creation process in the work information to the server 400, and the server 400 performs computationally intensive tasks such as multimodal semantic recognition, evaluation text generation and compliance verification, and returns the generated target evaluation text to the display device 200 for display.

[0117] In some embodiments, when the network connection is unavailable, the controller 250 automatically switches to local processing mode and uses the lightweight model built into the display device 200 to complete all the above processing steps, ensuring that the user can still be provided with interactive feedback function of the completed work in offline scenarios.

[0118] In this embodiment, cloud-based collaborative processing reduces the local computing load on the display device 200 and improves the response speed of evaluation generation. Simultaneously, the server 400 can utilize more powerful computing resources to run multimodal semantic models and evaluation generation models with larger parameter scales, further improving the accuracy of image detail recognition and the quality of evaluation text.

[0119] Figure 8 This is a schematic diagram illustrating the speech reading process provided in some embodiments of this application. For example... Figure 8As shown, in some embodiments, the controller 250 is also configured to convert the target evaluation text into speech reading data (S801), and to simultaneously display the text content of the target evaluation text and the playback status of the speech reading data on the display 260 (S802).

[0120] In some implementations, the controller 250 employs text-to-speech (TTS) technology to convert the target evaluation text into spoken data. For example, such as... Figure 9 As shown, while the controller 250 displays the text content of the target evaluation text on the display 260, it also plays the corresponding voice reading data through the audio output device 270. The target evaluation text is: "I see you drew a Chinese knot. The red color scheme is very eye-catching. I can see that you drew the weaving texture very carefully. Great job!" The evaluation is output synchronously through both text and voice channels, which is suitable for users such as children and elderly users who have difficulty reading text.

[0121] In some implementations, when the display device 200 plays the corresponding voice reading data through the audio output device 270, it also displays a voice playback progress bar or animated icon on the display 260 to prompt the user about the current playback status.

[0122] In some implementations, when invoking the TTS engine, the controller 250 sets speech synthesis parameters, including speech rate, pitch, and timbre, based on the creator's user profile. For example, when the creator is a child, a slower, higher-pitched child's voice is used to make the reading more natural and relatable; when the creator is an adult user, a moderately fast, natural-sounding adult voice is used. After the speech data is generated, the controller 250 plays it through the audio output device 270 and simultaneously displays speech playback controls on the display 260, including a playback progress bar, a pause button, and a volume adjustment button. The user can control the speech playback status through the user input interface 280, enabling free switching between text reading and speech listening modes. When the user pauses speech playback, the text content remains displayed, and the user can resume playback at any time; when the user turns off speech playback, the text content continues to be displayed, without affecting the user's reading and evaluation of the text.

[0123] Figure 10 This is a schematic diagram illustrating the process of generating a work version record for some embodiments of this application. For example... Figure 10As shown, in some embodiments, the controller 250 is further configured to: bind the target evaluation text to the corresponding creative work, generating a work version record (S1001). The work version record includes evaluation content, review items, and verification results. The work version record is then stored in a work version repository (S1002). For example, the work version record may include a thumbnail of the work image, the full text of the target evaluation text, details of the three-layer compliance verification results, a generation timestamp, and other information.

[0124] This disclosed embodiment enables full lifecycle management of creative works by binding compliance evaluations with the works for archiving and storage.

[0125] Figure 11 These are interactive diagrams illustrating the interaction methods based on the completion state of a work, provided in some embodiments of this application. For example... Figure 11 As shown, the user inputs a creation completion signal through the user input interface 280. After capturing the creation completion signal, the controller 250 locks the creation information (including the work image, work summary, and evidence of the creation process). The controller 250 parses the work information using a multimodal semantic model, extracts visible details and completion information of the work, and generates candidate evaluation text based on this. After compliance verification, if the candidate evaluation text passes the verification, it is pushed to the display 260 as the target evaluation text. If the verification fails, the candidate evaluation text is regenerated with rejection reason information until the verification passes.

[0126] In this embodiment, the user input interface 280 is responsible for front-end signal acquisition, converting user input signals such as touch, voice, and saved or static states into standardized creation completion signals and transmitting them to the controller 250. The controller 250 is responsible for data processing tasks such as locking artwork information, extracting multimodal features, generating evaluation text, and verifying compliance. The display 260 is responsible for the visual presentation of the final evaluation results, including the pop-up card display of the target evaluation text and the synchronized display of the voice playback control. Data transmission between modules is achieved through the internal bus of the display device 200, ensuring the real-time nature of the interaction process and data consistency.

[0127] Based on the display device 200 described above, some embodiments of this application also provide an interactive method based on the completion state of a work, which can be applied to the display device 200. The method includes the following steps: S401. Capture the creation completion signal input by the user through the user input interface.

[0128] S402. In response to the completion signal of creation, lock the work information corresponding to the created work.

[0129] S403. Analyze the artwork information and extract the visible details and completion information of the artwork from the artwork information.

[0130] S404. Generate candidate evaluation text based on the visible details and completion information of the work.

[0131] S405. Perform compliance verification on candidate evaluation texts.

[0132] S406. If the compliance verification result is unqualified, then the candidate evaluation text shall be regenerated based on the verification result.

[0133] S407. If the compliance verification result is qualified, the candidate evaluation text will be pushed to the display as the target evaluation text for display.

[0134] It should be noted that the interactive method based on the completed state of the work described in this application embodiment can adopt the same principle and implementation method as the above-mentioned display device 200 embodiment, and will not be described again here.

[0135] Similar parts between the embodiments provided in this application can be referred to mutually. The specific implementation methods provided above are only a few examples under the overall concept of this application and do not constitute a limitation on the scope of protection of this application. For those skilled in the art, any other implementation methods extended from the solution of this application without creative effort shall fall within the scope of protection of this application.

Claims

1. A display device, characterized in that, include: monitor; The user input interface is configured to collect user input signals indicating completion of creation; The controller is electrically connected to the display and the user input interface, respectively, and the controller is configured to: The creation completion signal input by the user is captured through the user input interface; In response to the creation completion signal, the artwork information corresponding to the created artwork is locked, and the artwork information includes the artwork image; The artwork information is analyzed, and visible details and completion information are extracted from the artwork information; the visible details are the visual content identified from the artwork image, and the completion information is the judgment result representing the area that does not need to be drawn and the degree of content completeness; Candidate evaluation text is generated based on the visible details and completion information of the work. Perform compliance checks on the candidate evaluation texts, including at least checking for any disabling intent to guide users to modify the work. If the compliance verification result is unqualified, the candidate evaluation text is regenerated based on the verification result. If the compliance verification result is qualified, the candidate evaluation text is pushed to the display as the target evaluation text for display.

2. The display device according to claim 1, characterized in that, The controller executes the work information corresponding to the locked creation work, specifically configured as follows: In response to the creation completion signal, freeze a snapshot of the created work; Extract the artwork image, artwork summary, and evidence of the creation process from the artwork snapshot as artwork information; The abstract of the work is a structured textual description of the content of the work, and the evidence of the creative process is the operation record of the entire creative process.

3. The display device according to claim 2, characterized in that, The controller is specifically configured to extract visible details and completion information of the artwork from the artwork information as follows: The artwork image, artwork summary, and creative process evidence are jointly identified using a multimodal semantic model, and the main subject, color features, background elements, composition layout, and brushstroke details are extracted as visible details of the artwork. The completion information of the artwork is determined based on whether there are areas to be added to the artwork image and the degree of completeness of the content.

4. The display device according to claim 3, characterized in that, The controller is specifically configured to generate candidate evaluation text based on the visible details and completion information of the artwork, as follows: Based on the main subject, color features, and background elements in the visible details of the artwork, generate observation and evaluation statements; Based on the brushstroke details and composition of the artwork, generate sincere words of encouragement. The observation and evaluation statements are combined with the sincere encouragement statements to form the candidate evaluation text.

5. The display device according to claim 4, characterized in that, The controller performs compliance checks and is specifically configured as follows: Verify whether the candidate evaluation text contains no less than a preset number of real, visible details to obtain a first verification result; The second verification result is obtained by verifying whether the candidate evaluation text contains encouraging statements that fit the content of the work and whether there are any templated phrases. The third verification result is obtained by verifying whether there are any statements in the candidate evaluation text that contain the intention to guide users to redraw, adjust or optimize their works. The compliance verification result is determined based on the first verification result, the second verification result, and the third verification result.

6. The display device according to claim 1, characterized in that, The controller is specifically configured to regenerate the candidate evaluation text based on the verification result as follows: When the compliance verification result is unqualified, the dimension information that failed the verification is summarized to generate rejection reason information; The rejection reason information and the rejected candidate evaluation text are input into the evaluation generation model to regenerate the updated candidate evaluation text. The compliance check is performed again on the updated candidate evaluation text.

7. The display device according to claim 1, characterized in that, The controller is also configured to: The target evaluation text is converted into speech reading data; The display simultaneously presents the text content of the target evaluation text and the playback status of the voice reading data.

8. The display device according to claim 1, characterized in that, The controller is also configured to: The target evaluation text is bound to the corresponding creative work to generate a work version record; the work version record includes evaluation content, review items, and verification results; The version records of the works are stored in the works version library.

9. The display device according to claim 1, characterized in that, The controller executes the creation completion signal captured from the user input via the user input interface, specifically configured as follows: Detect at least one of the following types of signals as the creation completion signal: Touch signals from users clicking the "Done" button, voice signals from users completing input, save status signals triggered by saving the work, and static status signals from prolonged periods without editing.

10. An interaction method based on the completion state of a work, characterized in that, include: Capture the creation completion signal input by the user through the user input interface; In response to the creation completion signal, the artwork information corresponding to the created artwork is locked, and the artwork information includes the artwork image; The artwork information is analyzed, and visible details and completion information are extracted from the artwork information; the visible details are the visual content identified from the artwork image, and the completion information is the judgment result representing the area that does not need to be drawn and the degree of content completeness; Candidate evaluation text is generated based on the visible details and completion information of the work. Perform compliance checks on the candidate evaluation texts, including at least checking for any disabling intent to guide users to modify the work. If the compliance verification result is unqualified, the candidate evaluation text is regenerated based on the verification result. If the compliance verification result is qualified, the candidate evaluation text will be displayed as the target evaluation text.

Images