Decorative equipment linkage method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a linkage method and device for decoration equipment, electronic equipment and a storage medium, and relates to the technical field of smart home, and the method comprises the steps: obtaining linkage data for the decoration equipment in the linkage control process of the decoration equipment, then carrying out the vectorization of the linkage data, and obtaining linkage data of the decoration equipment; the method comprises the steps of obtaining a corresponding scene feature vector according to the scene feature vector, then performing content prediction according to the scene feature vector to obtain visual content, and finally displaying the visual content according to a display parameter corresponding to the decorative equipment, thereby obtaining multi-modal linkage data and generating the corresponding visual content based on the linkage data. The content generation diversity and personalization are improved, the intelligence of the decoration equipment can be improved, the decoration equipment can present different contents according to the changed linkage data, the decoration effect is improved, and good visual experience is brought to a user.

Description

Technical Field

[0001] This invention relates to the field of smart home technology, and in particular to a method for linking decorative equipment, a device for linking decorative equipment, an electronic device, and a computer-readable storage medium. Background Technology

[0002] With the development of smart home technology, interconnecting and intelligentizing smart home devices has brought users numerous conveniences and upgraded experiences. This not only improves the ease of use and efficiency of smart home devices but also saves energy and costs, provides users with better security and protection, and enhances user comfort. In particular, smart home technology empowers decorative devices (such as lighting, curtains, walls, and decorations), enhancing the aesthetics and atmosphere of the home environment. However, for decorative devices, they still rely on manual user settings, lacking real-time capabilities, having low automation levels, and producing poor decorative effects. Summary of the Invention

[0003] The present invention provides a linkage method, device, electronic device, and computer-readable storage medium for decorative equipment, in order to solve or partially solve the problems of lack of real-time performance, low degree of automation, and poor decorative effect of decorative equipment.

[0004] This invention discloses a linkage method for decorative equipment, comprising:

[0005] Obtain linkage data for decorative equipment;

[0006] The linked data is vectorized to obtain the corresponding scene feature vector;

[0007] Based on the scene feature vector, content prediction is performed to obtain visual content;

[0008] The visual content is displayed according to the display parameters corresponding to the decorative device.

[0009] In some feasible implementations, the linkage data includes at least environmental data of the environment in which the decorative device is located, scene tags set by the user for the decorative device, and user preferences. The step of vectorizing the linkage data to obtain the corresponding scene feature vector includes:

[0010] The environmental data, scene labels, and user preferences are integrated to obtain the corresponding scene feature vector.

[0011] In some feasible implementations, the step of predicting content based on the scene feature vector to obtain visual content includes:

[0012] The scene feature vectors are used to construct content, resulting in low-resolution content.

[0013] The low-resolution content is used for detail optimization to obtain several candidate contents;

[0014] Visual content is selected from the candidate content.

[0015] In some feasible implementations, the step of using the scene feature vector to construct content and obtain low-resolution content includes:

[0016] Obtain the style label corresponding to the scene feature vector;

[0017] The scene feature vector and the style tag are used to construct content and obtain low-resolution content.

[0018] In some feasible implementations, the step of constructing content using the scene feature vector and the style tag to obtain low-resolution content includes:

[0019] The style weight vector corresponding to the style label is parsed from the scene feature vector;

[0020] The style labels are integrated according to the style weight vectors to obtain a comprehensive style feature vector;

[0021] Construct initial content corresponding to the scene feature vector;

[0022] The initial content is style-transferred using the comprehensive style feature vector to obtain low-resolution content with the style corresponding to the comprehensive style feature vector.

[0023] In some feasible implementations, the display parameters include at least a first resolution, color gamut, brightness information, and contrast information. Displaying the visual content according to the display parameters corresponding to the decorative device includes:

[0024] The visual content is displayed on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information.

[0025] In some feasible implementations, displaying the visual content on the decorative device according to at least one of the resolution, the color gamut, the brightness information, and the contrast information includes:

[0026] Obtain the second resolution of the visual content;

[0027] If the second resolution is less than or equal to the first resolution, the visual content is displayed on the decorative device according to the second resolution;

[0028] If the second resolution is greater than the first resolution, the second resolution is adjusted by an interpolation algorithm to at least reduce it to below the first resolution to obtain a corresponding third resolution, and the visual content is displayed on the decorative device according to the third resolution.

[0029] In some feasible implementations, displaying the visual content on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information includes:

[0030] Obtain the color data of the visual content and the color space of the decorative device;

[0031] The color data is converted according to the color space, and the first color temperature value of the visual content after color conversion and the second color temperature value of the environment in which the decorative device is located are obtained.

[0032] The first color temperature value is adjusted until the color temperature deviation between the first color temperature value and the second color temperature value is less than or equal to a preset threshold, and visual content is displayed on the decorative device based on the adjustment result.

[0033] In some feasible implementations, displaying the visual content on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information includes:

[0034] Obtain the light intensity of the environment in which the decorative equipment is located;

[0035] The brightness of the decorative device is adjusted to a preset brightness value according to the light intensity, and the visual content is displayed on the decorative device according to the preset brightness value.

[0036] In some feasible implementations, the linkage data includes at least scene tags set by the user for the decorative device, and displaying the visual content on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information includes:

[0037] The contrast information of the decorative device is adjusted to the target contrast information corresponding to the scene label, and the visual content is displayed on the decorative device according to the target contrast information.

[0038] This invention also discloses a linkage device for decorative equipment, comprising:

[0039] The data acquisition module is used to acquire linkage data for the decoration equipment;

[0040] The vectorization module is used to vectorize the linkage data to obtain the corresponding scene feature vectors.

[0041] The content prediction module is used to predict the content based on the scene feature vector to obtain the visual content.

[0042] The content display module is used to display the visual content according to the display parameters corresponding to the decorative device.

[0043] In some feasible implementations, the linkage data includes at least environmental data of the environment in which the decorative device is located, scene tags set by the user for the decorative device, and user preferences. The vectorization module is specifically used for:

[0044] The environmental data, scene labels, and user preferences are integrated to obtain the corresponding scene feature vector.

[0045] In some feasible implementations, the content prediction module is specifically used for:

[0046] The scene feature vectors are used to construct content, resulting in low-resolution content.

[0047] The low-resolution content is used for detail optimization to obtain several candidate contents;

[0048] Visual content is selected from the candidate content.

[0049] In some feasible implementations, the content prediction module is specifically used for:

[0050] Obtain the style label corresponding to the scene feature vector;

[0051] The scene feature vector and the style tag are used to construct content and obtain low-resolution content.

[0052] In some feasible implementations, the content prediction module is specifically used for:

[0053] The style weight vector corresponding to the style label is parsed from the scene feature vector;

[0054] The style labels are integrated according to the style weight vectors to obtain a comprehensive style feature vector;

[0055] Construct initial content corresponding to the scene feature vector;

[0056] The initial content is style-transferred using the comprehensive style feature vector to obtain low-resolution content with the style corresponding to the comprehensive style feature vector.

[0057] In some feasible implementations, the display parameters include at least a first resolution, color gamut, brightness information, and contrast information, and the content display module is specifically used for:

[0058] The visual content is displayed on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information.

[0059] In some feasible implementations, the content display module is specifically used for:

[0060] Obtain the second resolution of the visual content;

[0061] If the second resolution is less than or equal to the first resolution, the visual content is displayed on the decorative device according to the second resolution;

[0062] If the second resolution is greater than the first resolution, the second resolution is adjusted by an interpolation algorithm to at least reduce it to below the first resolution to obtain a corresponding third resolution, and the visual content is displayed on the decorative device according to the third resolution.

[0063] In some feasible implementations, the content display module is specifically used for:

[0064] Obtain the color data of the visual content and the color space of the decorative device;

[0065] The color data is converted according to the color space, and the first color temperature value of the visual content after color conversion and the second color temperature value of the environment in which the decorative device is located are obtained.

[0066] The first color temperature value is adjusted until the color temperature deviation between the first color temperature value and the second color temperature value is less than or equal to a preset threshold, and visual content is displayed on the decorative device based on the adjustment result.

[0067] In some feasible implementations, the content display module is specifically used for:

[0068] Obtain the light intensity of the environment in which the decorative equipment is located;

[0069] The brightness of the decorative device is adjusted to a preset brightness value according to the light intensity, and the visual content is displayed on the decorative device according to the preset brightness value.

[0070] In some feasible implementations, the linkage data includes at least scene tags set by the user for the decorative device, and the content display module is specifically used for:

[0071] The contrast information of the decorative device is adjusted to the target contrast information corresponding to the scene label, and the visual content is displayed on the decorative device according to the target contrast information.

[0072] This invention also discloses an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0073] The memory is used to store computer programs;

[0074] When the processor executes a program stored in the memory, it implements the method described in the embodiments of the present invention.

[0075] This invention also discloses a computer-readable storage medium storing instructions that, when executed by one or more processors, cause the processors to perform the methods described in this invention.

[0076] The embodiments of the present invention have the following advantages:

[0077] In this embodiment of the invention, during the linkage control of decorative equipment, linkage data for the decorative equipment is acquired, then the linkage data is vectorized to obtain corresponding scene feature vectors, and then content prediction is performed based on the scene feature vectors to obtain visual content. Finally, the visual content is displayed according to the display parameters corresponding to the decorative equipment. Thus, by acquiring multimodal linkage data and generating corresponding visual content based on the linkage data, not only is the diversity and personalization of content generation improved, but the intelligence of the decorative equipment is also enhanced, enabling it to present different content according to changing linkage data, thereby improving the decorative effect and bringing a good visual experience to the user. Attached Figure Description

[0078] Figure 1 This is a flowchart illustrating the steps of a linkage method for a decorative device provided in an embodiment of the present invention;

[0079] Figure 2 This is a flowchart illustrating the smart home system provided in this embodiment of the invention;

[0080] Figure 3 This is a structural block diagram of a linkage device for a decorative equipment provided in an embodiment of the present invention. Detailed Implementation

[0081] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0082] As an example, interior decorative paintings are usually displayed in electronic frames. The updating of the content relies on manual presets, lacking real-time and automation. Even if the content can be generated using AI (Artificial Intelligence) technology, the limited application scope prevents it from being integrated with smart scenarios, resulting in relatively simple content that cannot be personalized or contextualized.

[0083] In this invention, during the linkage control of decorative equipment, linkage data for the decorative equipment is acquired, then vectorized to obtain corresponding scene feature vectors. Content prediction is then performed based on these scene feature vectors to obtain visual content. Finally, the visual content is displayed according to the display parameters of the decorative equipment. By acquiring multimodal linkage data and generating corresponding visual content based on it, not only is the diversity and personalization of content generation improved, but the intelligence of the decorative equipment is also enhanced, enabling it to present different content based on changing linkage data, thus improving the decorative effect and providing users with a better visual experience.

[0084] Reference Figure 1 The diagram illustrates a flowchart of a linkage method for a decorative device provided in an embodiment of the present invention, which may specifically include the following steps:

[0085] Step 101: Obtain linkage data for the decorative equipment;

[0086] In this embodiment of the invention, the decorative device can be a terminal (such as a smart hanging picture terminal) for presenting corresponding image content. It can be configured with a graphical user interface, a control unit, a communication unit, and an actuator. The graphical user interface can be used to display the corresponding image content; the control unit can be used to implement localized AI inference and corresponding data processing; the communication unit can be used for data communication, such as receiving control commands sent by the user terminal and executing operations corresponding to the control commands; the actuator can include a brightness adjustment module and an image scaling engine. The brightness adjustment module can adjust the brightness of the image, and the image scaling engine can adapt the frame ratio, etc. This invention does not limit these aspects.

[0087] In some feasible implementation manners, a smart home system including a decoration device can be constructed. The smart home system can include a smart scene perception module, a content generation module, a decoration device, a central control platform, etc. Among them, the smart scene perception module can include various sensors and user terminals, etc., for collecting linkage data, etc.; the content generation module can be a module constructed based on a deep learning model for content prediction; the central control platform can be used for real-time processing of sensor data, data storage, updating model parameters, and remote OTA (Over-The-Air) upgrade, etc. Through this smart home system, personalized, scenario-based, and intelligent linkage control of the decoration device can be achieved to provide a good visual experience for users.

[0088] Optionally, during the process of linkage control of the decoration device, the linkage data can include multi-modal data, such as environmental data collected by environmental sensors, behavior data collected by human presence sensors, and preference data set by the user through the user terminal, etc. The present invention does not limit this.

[0089] Step 102: Vectorize the linkage data to obtain a corresponding scene feature vector;

[0090] In the embodiment of the present invention, after obtaining the corresponding linkage data, all the linkage data can be vectorized and integrated into a corresponding scene feature vector. Different modalities corresponding to features, such as environmental features, behavior features, and user features, etc., can be represented by this scene feature vector, so as to perform content prediction based on this scene feature vector. Among them, the linkage data at least includes environmental data of the environment where the decoration device is located, scene labels set by the user for the decoration device (used to reflect the user's behavior characteristics), and user preferences. Then, the environmental data, scene labels, and user preferences are integrated to obtain a corresponding scene feature vector.

[0091] For example, the sensor group can report the collected environmental parameters regularly (such as once every 500 ms, etc.) or dynamically (such as reporting the changed data when the environmental data changes, etc.). Combining the scene labels triggered by the user (such as "morning wake-up mode", "meeting mode", etc.) and the preferences (color preference, style preference) set by the user in the APP (Application), etc., based on the collected data, a composite scene feature vector F = [E, B, P] can be generated at the edge node. E is the environmental feature, B is the behavior feature, and P is the preference feature. Different modalities corresponding to features are represented by this scene feature vector.

[0092] In some examples, for the vectorization process, it can be as follows:

[0093] 1. Vectorization process of environmental features:

[0094] Optionally, environmental features may include continuous values ​​such as temperature, humidity, and light intensity, as well as some binary or categorical values ​​(such as whether someone is present). Different vectorization methods can be used depending on the data format. For example, for continuous values, the original numerical values ​​can be used directly; for binary values, 0 and 1 can be used; and for categorical values, one-hot encoding can be used.

[0095] Example:

[0096] Temperature: 24℃; Humidity: 50%; Light intensity: 300 lux; Is anyone present: Yes.

[0097] Correspondingly, the vectorized environment feature vector E can be expressed as:

[0098] E = [24, 50, 300, 1]

[0099] 2. Behavioral Feature Vectorization

[0100] Optionally, behavioral characteristics may include user-triggered scene tags and related behavioral patterns, which can be represented by timestamps, event types, and other relevant parameters.

[0101] Example:

[0102] Scene tag: "Morning Mode"; Time: 7:00 AM;

[0103] Lighting preference: soft natural light; Equipment used: coffee machine.

[0104] The following methods can be used to vectorize behavioral features:

[0105] Scene tag: Use one-hot encoding.

[0106] Time: This can be converted to hours (e.g., 7:00 AM is 7).

[0107] Light preference: Use predefined numerical values ​​(e.g., soft natural light = 1, strong light = 2).

[0108] Device usage: Uses a binary value to indicate whether the device is turned on.

[0109] Assuming that corresponding mapping rules are pre-set, such as:

[0110] Scene tag: Morning mode -> [1, 0, 0];

[0111] Lighting preference: Soft natural light -> 1;

[0112] Coffee machine turned on -> 1.

[0113] The vectorized behavioral feature vector B can then be represented as:

[0114] B = [1, 0, 0, 7, 1, 1]

[0115] 3. Personalized Preference Vectorization

[0116] Personalized preferences typically include color preferences, style preferences, and special needs. This information can also be represented by predefined numerical values ​​or one-hot encoding.

[0117] Example:

[0118] Color preference: warm colors; Style preference: modern minimalist style; Special needs: sufficient but not glaring light for reading at night, etc.

[0119] Assuming that corresponding mapping rules are pre-set, such as:

[0120] Color preference: Warm colors -> 1;

[0121] Style preference: Modern minimalist style -> [1, 0, 0];

[0122] Special requirement: Suitable for nighttime reading -> 1.

[0123] The vectorized personalized preference vector P can be represented as:

[0124] P = [1, 1, 0, 0, 1]

[0125] Combined to form a composite scene feature vector

[0126] Combining the above three parts forms the final composite scene feature vector F: F = [E, B, P], that is, F = [24, 50, 300, 1, 1, 0, 0, 7, 1, 1, 1, 1, 0, 0, 1].

[0127] Through the above process, multimodal data can be vectorized into scene feature vectors, which can then be used to predict content, obtain content associated with multiple dimensions, enrich the diversity of content, and enhance the scene linkage between smart home scenarios.

[0128] Step 103: Perform content prediction based on the scene feature vector to obtain visual content;

[0129] Once the corresponding scene feature vector is constructed, content prediction can be performed based on the scene feature vector to obtain the corresponding visual content. This enables the generation of corresponding visual content in real time based on the environmental characteristics of the smart home environment, user behavior characteristics, and personalized preferences. This achieves intelligent, scenario-based, and personalized home decoration, improves user experience, and reduces the cost of replacing decorations based on the real-time generated visual content.

[0130] In some feasible implementations, a two-stage generation architecture can be used to generate visual content. First, the scene feature vectors are used to construct content and obtain low-resolution content. Then, the low-resolution content is used for detail optimization to obtain several candidate contents. Finally, the visual content is selected from the candidate contents. Thus, the two-stage generation architecture can effectively improve the quality and diversity of generated content, optimize the use of computing resources, and quickly generate low-resolution content first, followed by detail optimization to improve the content's refinement and provide users with a good visual experience.

[0131] In some feasible examples, during the low-resolution (e.g., 256*256) generation process, the UNet backbone network can be used, taking the scene feature vector F as a conditional input and guiding the generation process through a cross-attention layer to obtain the corresponding low-resolution content. During the super-resolution (e.g., 3840*2160) enhancement process, based on the low-resolution content, a cascaded Upsampler module combined with local detail attention can be used to improve image resolution and detail. Thus, through a two-stage generation architecture, the quality and diversity of generated content can be effectively improved, the use of computing resources can be optimized, and low-resolution content can be generated quickly first, followed by a detail optimization process to improve the content's refinement, providing users with a good visual experience.

[0132] Furthermore, various art styles (such as Impressionism, ink painting, etc.) can be pre-trained based on the VGG-19 network. By extracting the feature information corresponding to different art styles, style references can be used. For example, feature statistics (mean / variance, etc.) corresponding to various art styles can be extracted as corresponding style parameters. In the VGG-19 network (Visual GeometryGroup-19layers, a 19-layer deep convolutional neural network), each art style (such as Impressionism, ink painting) will extract feature maps at different levels (such as convolutional layers 1 to 5), and calculate the mean and variance of these feature maps as style references.

[0133] For example: Suppose there are three art styles (Impressionism, ink painting, cyberpunk, etc.), then the characteristic statistics of each style can be expressed as:

[0134] style_reference={

[0135] 'impressionist':{'mean':[0.7,0.6,0.5],'var'[0.1,0.1,0.1]},

[0136] 'ink_wash':{'mean'[0.4,0.3,0.2],'var'[0.2,0.2,0.2]},

[0137] 'cyberpunk':{'mean'[0.9,0.8,0.7],'var'[0.05,0.05,0.05]}

[0138] }

[0139] After constructing the corresponding artistic style, adaptive style fusion can be performed during the content prediction process. Specifically, style tags corresponding to the scene feature vector can be obtained, and then the scene feature vector and the style tags can be used to construct content to obtain low-resolution content. This allows for the dynamic mixing of multiple style features, enabling the generated content to match user preferences and achieving personalized content prediction.

[0140] In a specific viewpoint, the style weight vector corresponding to the style tag can be parsed from the scene feature vector. Then, the style tags are integrated according to the style weight vector to obtain a comprehensive style feature vector. Then, initial content corresponding to the scene feature vector is constructed. Finally, the comprehensive style feature vector is used to perform style transfer on the initial content to obtain low-resolution content with the style corresponding to the comprehensive style feature vector. This dynamically mixes multiple style features, making the generated content match user preferences and realizing personalized content prediction.

[0141] For example, in the process of adaptive style transfer, multiple art styles can be dynamically mixed based on scene characteristics and user preferences. Corresponding functionalities could include:

[0142] Style weight calculation: Extract the style weight vector from the scene features.

[0143] Style feature fusion: Generate comprehensive style features through weighted summation.

[0144] Style transfer application: Generate the final image using a pre-trained style transfer network.

[0145] To achieve the above process, the corresponding content prediction process can be implemented using the following code:

[0146] def adaptive_style_transfer(content_img,style_params):

[0147] #style_params: Style weight vector parsed from scene features F

[0148] target_style=torch.zeros_like(style_reference['impressionist'])

[0149] for i,style_name in enumerate(style_names):

[0150] target_style+=style_params[i]*style_reference[style_name]

[0151] return style_transfer_network(content_img,target_style)

[0152] Here, content_img is the content image (the original image output by the AI ​​generation module); style_params is the style weight vector parsed from the scene feature vector F (such as the style proportion in user preferences); target_style is the weighted combination of target style features (such as Impressionism style accounting for 70%, ink painting style accounting for 30%, etc.). The final returned image is a new image after being processed by the style transfer network (integrating decorative effects of multiple styles), so that the corresponding target content can be obtained based on the subsequent detail enhancement process.

[0153] In some examples, taking image generation as an example, the simplified conditional Diffusion model structure, when combined with style tags, can be as follows:

[0154] class ConditionalDiffusion(nn.Module):

[0155] def__init__(self,img_size,feature_dim,num_classes):

[0156] super().__init__()

[0157] self.unet = UNet(

[0158] in_channels=3,

[0159] out_channels = 3,

[0160] context_dim = feature_dim, # Dimension of scene feature vector

[0161] num_classes = num_classes # Number of style tags )

[0163] self.scheduler=DDPMScheduler(num_train_timesteps=1000)

[0164] def forward(self,x,t,context,style_label=None):

[0165] #x: Input noisy image

[0166] #t: number of diffusion steps

[0167] #context: Scene feature vector F

[0168] Return self.unet(x,t,encoder_hidden_states=context,class_labels=style_label)

[0169] The code above defines an image generation module based on the Diffusion model, capable of generating high-quality images according to scene features and style preferences. Its core functions include:

[0170] UNet architecture: The backbone network for image generation.

[0171] DDPM Scheduler (Denoising Diffusion Probabilistic Model Scheduler): Controls the number of Diffusion training steps.

[0172] Scene feature vector F: serves as a conditional input, guiding the generation process.

[0173] Style tag: Used to specify the target art style.

[0174] Through the above process, the content prediction model can first generate a set of candidate images {I1,I2,...,In}, and then use the NMS (Non-Maximum Suppression) algorithm to filter the optimal image I*, thereby selecting the best one from multiple candidate contents, reducing redundant content and improving the accuracy and efficiency of the prediction results.

[0175] Step 104: Display the visual content according to the display parameters corresponding to the decorative device.

[0176] Once the visual content is determined, the display parameters corresponding to the decorative equipment can be obtained. Then, based on these display parameters, the visual content is displayed on the decorative equipment. By acquiring multimodal linkage data and generating corresponding visual content based on this data, not only is the diversity and personalization of content generation enhanced, but the intelligence of the decorative equipment is also improved. This allows the equipment to present different content based on changing linkage data, improving the decorative effect and providing users with a better visual experience. Furthermore, by combining the display parameters of the decorative equipment itself, normal display on the visual equipment is ensured, enhancing the flexibility and versatility of content display.

[0177] In some feasible implementations, the display parameters include at least a first resolution, color gamut, brightness information, and contrast information, so that the visual content can be displayed on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information.

[0178] Regarding resolution, a second resolution of the visual content can be obtained, and then the first resolution can be compared with the second resolution. If the second resolution is less than or equal to the first resolution, the visual content is displayed on the decorative device according to the second resolution. If the second resolution is greater than the first resolution, the second resolution is adjusted by an interpolation algorithm to at least reduce it to below the first resolution to obtain the corresponding third resolution, and the visual content is displayed on the decorative device according to the third resolution.

[0179] For color gamut, the color data of the visual content and the color space of the decorative device can be obtained. Then, the color data can be converted according to the color space, and the first color temperature value of the visual content after color conversion and the second color temperature value of the environment in which the decorative device is located can be obtained. Then, the first color temperature value can be adjusted until the color temperature deviation between the first color temperature value and the second color temperature value is less than or equal to a preset threshold. Based on the adjustment result, the visual content is displayed on the decorative device.

[0180] Regarding brightness information, the light intensity of the environment in which the decorative device is located can be obtained, and then the brightness information of the decorative device can be adjusted to a preset brightness value according to the light intensity, and the visual content can be displayed on the decorative device according to the preset brightness value.

[0181] Regarding contrast information, the contrast information of the decorative device can be adjusted to the target contrast information corresponding to the scene label, and the visual content can be displayed on the decorative device according to the target contrast information.

[0182] For example, in the adaptive content generation phase, the central control platform needs to adapt the optimal image I* output by the AI ​​generation module with the hardware parameters of the smart wall hanging terminal. This adaptation process may include the following steps:

[0183] ① Resolution adaptation

[0184] Assuming the smart wall hanging terminal has a screen resolution of 300 PPI or higher (such as 4K flexible OLED (Organic Light-Emitting Diode)), the candidate image output by the AI ​​generation module may be 3840×2160 or higher. If the terminal supports this resolution, it is output directly; otherwise, it is reduced to a resolution supported by the terminal (e.g., 2560×1440) through an interpolation algorithm (such as bicubic interpolation).

[0185] ② Color gamut conversion

[0186] Assuming the image output by the AI ​​generation module is in sRGB format, while the smart wall hanging terminal may use DisplayP3 or other professional color spaces, color gamut conversion can be completed through a color rendering index matching algorithm to ensure that the color temperature of the image deviates from the ambient light color temperature by ≤100K and minimizes the loss of color accuracy.

[0187] ③Brightness adjustment

[0188] The screen brightness is dynamically adjusted based on data from the light intensity sensor. For example, when the current ambient light is 600 lux, it is automatically adjusted to 400 nits (a unit of brightness).

[0189] ④ Contrast optimization

[0190] Adjust the contrast parameter (e.g., 1000:1) based on user preferences (e.g., low-contrast blue-toned images) and behavioral data (e.g., detected sleep patterns) in the scene feature vector.

[0191] Furthermore, in the above process, taking a single adjustment dimension as an example for illustration, it can be understood that in practical applications, adaptive adjustments can be made based on one, two, or more dimensions to match the display of visual content with the decorative equipment, the environment in which the decorative equipment is located, and user preferences, thereby improving the automation, intelligence, and personalization of content display.

[0192] Through the above adaptive adjustment process, combined with the display parameters of the decorative equipment itself, normal display on the visual equipment is ensured, and the flexibility and versatility of content display are improved.

[0193] It should be noted that the embodiments of the present invention include, but are not limited to, the examples described above. It is understood that those skilled in the art can make further settings according to actual needs under the guidance of the ideas in the embodiments of the present invention, and the present invention does not limit such settings.

[0194] In this embodiment of the invention, during the linkage control of decorative equipment, linkage data for the decorative equipment is acquired, then the linkage data is vectorized to obtain corresponding scene feature vectors, and then content prediction is performed based on the scene feature vectors to obtain visual content. Finally, the visual content is displayed according to the display parameters corresponding to the decorative equipment. Thus, by acquiring multimodal linkage data and generating corresponding visual content based on the linkage data, not only is the diversity and personalization of content generation improved, but the intelligence of the decorative equipment is also enhanced, enabling it to present different content according to changing linkage data, thereby improving the decorative effect and bringing a good visual experience to the user.

[0195] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the following examples are provided for illustrative purposes:

[0196] In some examples, refer to Figure 2 The diagram illustrates a process flow chart of a smart home system provided in an embodiment of the present invention, which may include:

[0197] 1. Intelligent Scene Perception Module: Composed of an environmental sensor group (such as light sensor, temperature and humidity sensor, etc.), a human presence sensor (such as infrared sensor, camera, etc.), and a user input terminal (such as mobile APP, voice interaction device, etc.), used to collect: Environmental data: light intensity / color temperature, temperature and humidity; Behavioral data: user location, activity status (such as reading, watching movies, sleeping); Preference data: user-preset art style (realistic / abstract), color preference, theme keywords, etc.

[0198] 2. AI image generation module, built based on a deep learning model, includes:

[0199] Data preprocessing unit: Normalizes the data from the perception module to generate scene feature vectors.

[0200] Generative model: Employs a cascaded Diffusion model, supports multi-condition input (scene features + user preferences), and outputs RGB images with a resolution ≥3840×2160.

[0201] Style Transfer Engine: Built-in VGG style transfer network, enabling real-time transfer of 10+ preset styles such as Impressionism and Cyberpunk.

[0202] 3. Smart picture hanging terminal, hardware configuration includes:

[0203] Display unit: 4K flexible OLED screen or electronic paper display (resolution above 300 PPI).

[0204] Control unit: integrates an ARM processor and an NPU computing chip, supporting local AI inference.

[0205] Communication module: Supports Wi-Fi 6, Bluetooth 5.2 and Zigbee 3.0 protocols.

[0206] Actuators: Automatic dimming module (supports 0-100% brightness adjustment), image scaling engine (adapts to frame ratio)

[0207] 4. The central control platform adopts a collaborative architecture of edge computing and cloud computing:

[0208] Edge nodes: Deploy scene decision-making algorithms, process sensor data in real time, and generate input parameters for the generative model (edge ​​nodes can be routers with integrated edge computing modules, or home hosts, home gateways, and other devices).

[0209] Cloud server: Stores user preference database, updates AI model parameters, and supports remote OTA upgrades.

[0210] For example, taking image generation as an example, during the linkage process of the smart wall hanging terminal, the intelligent scene perception module is responsible for collecting relevant data (such as user preference settings / text input in the user input terminal, activity status collected by the human presence sensor, and data such as light, temperature, and humidity collected by the environmental sensor). Then, the central control platform can generate corresponding images based on the collected data through the AI ​​image generation module, and then transmit the images to the smart wall hanging terminal. The smart wall hanging terminal displays the corresponding images. By acquiring multimodal linkage data and generating corresponding visual content based on the linkage data, not only is the diversity and personalization of content generation improved, but the intelligence of decorative equipment can also be improved, enabling it to present different content according to changing linkage data, thereby improving the decorative effect and bringing a good visual experience to users.

[0211] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0212] Reference Figure 3 The diagram shows a structural block diagram of a linkage device for a decorative equipment provided in an embodiment of the present invention, which may specifically include the following modules:

[0213] Data acquisition module 301 is used to acquire linkage data for the decoration equipment;

[0214] The vectorization module 302 is used to vectorize the linkage data to obtain the corresponding scene feature vector;

[0215] Content prediction module 303 is used to predict content based on the scene feature vector to obtain visual content;

[0216] The content display module 304 is used to display the visual content according to the display parameters corresponding to the decorative device.

[0217] In some feasible implementations, the linkage data includes at least environmental data of the environment in which the decorative device is located, scene tags set by the user for the decorative device, and user preferences. The vectorization module 302 is specifically used for:

[0218] The environmental data, scene labels, and user preferences are integrated to obtain the corresponding scene feature vector.

[0219] In some feasible implementations, the content prediction module 303 is specifically used for:

[0220] The scene feature vectors are used to construct content, resulting in low-resolution content.

[0221] The low-resolution content is used for detail optimization to obtain several candidate contents;

[0222] Visual content is selected from the candidate content.

[0223] In some feasible implementations, the content prediction module 303 is specifically used for:

[0224] Obtain the style label corresponding to the scene feature vector;

[0225] The scene feature vector and the style tag are used to construct content and obtain low-resolution content.

[0226] In some feasible implementations, the content prediction module 303 is specifically used for:

[0227] The style weight vector corresponding to the style label is parsed from the scene feature vector;

[0228] The style labels are integrated according to the style weight vectors to obtain a comprehensive style feature vector;

[0229] Construct initial content corresponding to the scene feature vector;

[0230] The initial content is style-transferred using the comprehensive style feature vector to obtain low-resolution content with the style corresponding to the comprehensive style feature vector.

[0231] In some feasible implementations, the display parameters include at least a first resolution, color gamut, brightness information, and contrast information, and the content display module 304 is specifically used for:

[0232] The visual content is displayed on the decorative device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information.

[0233] In some feasible implementations, the content display module 304 is specifically used for:

[0234] Obtain the second resolution of the visual content;

[0235] If the second resolution is less than or equal to the first resolution, the visual content is displayed on the decorative device according to the second resolution;

[0236] If the second resolution is greater than the first resolution, the second resolution is adjusted by an interpolation algorithm to at least reduce it to below the first resolution to obtain a corresponding third resolution, and the visual content is displayed on the decorative device according to the third resolution.

[0237] In some feasible implementations, the content display module 304 is specifically used for:

[0238] Obtain the color data of the visual content and the color space of the decorative device;

[0239] The color data is converted according to the color space, and the first color temperature value of the visual content after color conversion and the second color temperature value of the environment in which the decorative device is located are obtained.

[0240] The first color temperature value is adjusted until the color temperature deviation between the first color temperature value and the second color temperature value is less than or equal to a preset threshold, and visual content is displayed on the decorative device based on the adjustment result.

[0241] In some feasible implementations, the content display module 304 is specifically used for:

[0242] Obtain the light intensity of the environment in which the decorative equipment is located;

[0243] The brightness of the decorative device is adjusted to a preset brightness value according to the light intensity, and the visual content is displayed on the decorative device according to the preset brightness value.

[0244] In some feasible implementations, the linkage data includes at least scene tags set by the user for the decorative device, and the content display module 304 is specifically used for:

[0245] The contrast information of the decorative device is adjusted to the target contrast information corresponding to the scene label, and the visual content is displayed on the decorative device according to the target contrast information.

[0246] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0247] In addition, this invention also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described linkage method embodiment of the decorative device and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0248] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described linkage method embodiment for the decorative device and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0249] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0250] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, EEPROM, Flash, and eMMC, etc.) containing computer-usable program code.

[0251] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0252] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0253] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0254] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0255] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0256] The above provides a detailed description of a linkage method and a linkage device for a decorative device provided by the present invention. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A linkage method of a decorating apparatus, characterized by, The method comprises: acquiring linkage data of a decoration device; vectorizing the linkage data to obtain a corresponding scene feature vector; performing content prediction according to the scene feature vector to obtain visual content; displaying the visual content according to display parameters corresponding to the decoration device.

2. The method of claim 1, wherein, The linkage data at least comprises environment data of an environment where the decoration device is located, a scene label set by a user for the decoration device, and user preferences, and the vectorizing the linkage data to obtain a corresponding scene feature vector comprises: integrating the environment data, the scene label, and the user preferences to obtain a corresponding scene feature vector.

3. The method of claim 1, wherein, The performing content prediction according to the scene feature vector to obtain visual content comprises: performing content construction using the scene feature vector to obtain low-resolution content; performing detail optimization using the low-resolution content to obtain a plurality of candidate contents; selecting visual content from the plurality of candidate contents.

4. The method of claim 3, wherein, The performing content construction using the scene feature vector to obtain low-resolution content comprises: acquiring a style label corresponding to the scene feature vector; performing content construction using the scene feature vector and the style label to obtain low-resolution content.

5. The method of claim 4, wherein, The performing content construction using the scene feature vector and the style label to obtain low-resolution content comprises: analyzing a style weight vector corresponding to the style label from the scene feature vector; integrating each style label according to each style weight vector to obtain a comprehensive style feature vector; constructing initial content corresponding to the scene feature vector; performing style migration on the initial content using the comprehensive style feature vector to obtain low-resolution content with a style corresponding to the comprehensive style feature vector.

6. The method of claim 1, wherein, The display parameters at least comprise a first resolution, a color gamut, luminance information, and contrast information, and the displaying the visual content according to the display parameters corresponding to the decoration device comprises: displaying the visual content on the decoration device according to at least one of the first resolution, the color gamut, the luminance information, and the contrast information.

7. The method of claim 6, wherein, The displaying the visual content on the decoration device according to at least one of the resolution, the color gamut, the luminance information, and the contrast information comprises: acquiring a second resolution of the visual content; if the second resolution is less than or equal to the first resolution, displaying the visual content on the decoration device according to the second resolution; if the second resolution is greater than the first resolution, adjusting the second resolution by an interpolation algorithm to at least reduce to be lower than the first resolution to obtain a corresponding third resolution, and displaying the visual content on the decoration device according to the third resolution.

8. The method of claim 6, wherein, The displaying the visual content on the decoration device according to at least one of the first resolution, the color gamut, the luminance information, and the contrast information comprises: acquiring color data of the visual content and a color space of the decoration device; The color data is converted according to the color space, and a first color temperature value of the visual content after color conversion and a second color temperature value of an environment where the decoration device is located are obtained; The first color temperature value is adjusted until a color temperature deviation between the first color temperature value and the second color temperature value is less than or equal to a preset threshold, and visual content is displayed on the decoration device based on an adjustment result.

9. The method of claim 6, wherein, The visual content is displayed on the decoration device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information, including: An illumination intensity of an environment where the decoration device is located is obtained; The brightness information of the decoration device is adjusted to a preset brightness value according to the illumination intensity, and the visual content is displayed on the decoration device according to the preset brightness value.

10. The method of claim 6, wherein, The linkage data at least includes a scene label set by a user for the decoration device, and the visual content is displayed on the decoration device according to at least one of the first resolution, the color gamut, the brightness information, and the contrast information, including: The contrast information of the decoration device is adjusted to target contrast information corresponding to the scene label, and the visual content is displayed on the decoration device according to the target contrast information.

11. A linkage for a decorating apparatus, characterized by including: A data acquisition module is configured to acquire linkage data for a decoration device; A vectorization module is configured to vectorize the linkage data to obtain a corresponding scene feature vector; A content prediction module is configured to predict content according to the scene feature vector to obtain visual content; A content display module is configured to display the visual content according to display parameters corresponding to the decoration device.

12. An electronic device, comprising: The processor, the communication interface, and the memory complete communication with each other through the communication bus; The memory is configured to store a computer program; The processor is configured to execute the program stored on the memory to implement the method of any one of claims 1-11.

13. A computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the method of any one of claims 1-11.

Images