63 results about "Drop shadow" patented technology

The invention discloses an immersive design experience method and system based on VR, and relates to the technical field of virtual reality, and the method comprises the steps: monitoring the position and attitude data of a dynamic object in a scene in real time between continuous rendering frames of the virtual reality scene, the position and attitude data comprising a three-dimensional coordinate, a rotation angle and a scaling; and when it is monitored that the position posture data of the dynamic object changes, determining a first projection area corresponding to the dynamic object in a previous frame rendering cache in a scene based on the position posture data before the change, and defining the first projection area as a shadow erasing area. According to the method, by dynamically monitoring the position and posture change of the object, accurately delimiting the shadow erasing area and the reconstruction area, and only executing shadow clearing and generation calculation on the local area, invalid operation is greatly reduced, calculation complexity is remarkably reduced, rendering efficiency is improved, and shadow updating delay and picture lagging are avoided.

The application discloses a sketch representation enhancement method and system based on self-supervised learning, in the image and text feature extraction stage, the sketch is augmented and transformed, the text box content in the sketch is recognized, and the text content and the sketch are encoded respectively to obtain text features and image features; in the sketch representation enhancement stage under the guidance of the text, the text features are used as the basis, and a guide attention unit is applied to enhance the image features; in the contrast self-supervised learning stage, the original image and the augmented sketch enhancement features are mapped into a low-dimensional vector space through a projection function, and the loss is calculated by the low-dimensional vector and the model is optimized.

1. The name of the design product: projector. 2. The use of the design product: for the projection and play of image, video and the output of audio. 3. The design points of the design product: in shape. 4. The picture or photo that can best show the design points: perspective view 1.

This invention provides a zero-shot anomaly detection method and system based on bidirectional alignment enhancement. The method includes: achieving direct interaction between global and local features through the cross-attention module CAGL; extracting fine-grained local features through the region token alignment module RegionAlign; and projecting global visual semantics onto the text embedding space through the text embedding enhancement module TextAug to generate domain-aware cue words. This invention solves the technical problems of poor generalization ability, neglect of local spatial details, and difficulty in hierarchical modeling of multi-scale visual anomalies.

1. The name of the design product: the front frame part of a projector. 2. The use of the design product: the product as a whole is a projector for projecting video, images, etc.; the part to be protected is the front frame part of the projector. 3. The design points of the design product: the shape shown by solid lines in the perspective view. 4. The picture or photo that best indicates the design points: the perspective view. 5. Other circumstances that need to be explained: the design is a partial design, the part shown by solid lines is the part to be protected as a partial design, and the dashed line is only a line indicating the boundary between the part to be protected as a partial design and other parts. The part other than the part marked in gray in the reference figure is the part to be protected.

1. Name of the product in this design: Desktop Ornament (Media-Free Holographic Opera Culture D). 2. Application of this design: This design is intended for use in devices for media-free holographic projection and sound playback. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model.

1. Name of the product in this design: Elevator status indicator with dynamic graphical user interface for elevator status. 2. Purpose of this design: To display the elevator status on the side of the lobby and project the elevator status onto the ground. 3. The key design features of this product are: the shape of the elevator status indicator and the content and layout of the graphical user interface displayed in the elevator status indicator. 4. The image or photograph that best illustrates the key design points: Design 1 3D view. 5. Design 1 is designated as the basic design. 6. Purpose of the graphical user interface: This graphical user interface is used to display the elevator status. Starting from the main view of Design 1, when the elevator is descending, the interface changes to the Design 1 state diagram; starting from the main view of Design 2, when the elevator is ascending, the interface changes to the Design 2 state diagram; starting from the main view of Design 3, when the elevator door opens, the interface changes to the Design 3 state diagram; starting from the main view of Design 4, when the elevator is overloaded, the interface changes to the Design 4 state diagram; starting from the main view of Design 5, when the elevator is fully loaded, the interface changes to the Design 5 state diagram; starting from the main view of Design 6, when the elevator is under maintenance, the interface changes to the Design 6 state diagram; starting from the main view of Design 7, after changing the wallpaper, the interface changes to the Design 7 state diagram; starting from the main view of Design 8, when it rains, the interface changes to the Design 8 state diagram; starting from the main view of Design 9, at night, the interface changes to the Design 9 state diagram; starting from the main view of Design 10, when an earthquake warning is issued, the interface changes to the Design 10 state diagram. The characters or patterns in the interface are variable. 7. Other situations requiring explanation: The front, rear, left, right, top, bottom, and perspective views of Designs 2 to 10 correspond to the same as those of Design 1. Therefore, the front, rear, left, right, top, bottom, and perspective views of Designs 2 to 10 are omitted. Design 1, reference figure 1 shows the system update status; Design 1, reference figure 2 shows the remote control status; Design 1, reference figure 3 shows the elevator descending status; Design 1, reference figure 4 shows the passenger arrival status when the elevator is descending; Design 1, reference figure 5 shows the elevator guidance status when the elevator is descending; Design 2, reference figure shows the elevator ascending and about to arrive status; Design 5, reference figure shows the elevator fully loaded status; Design 6, reference figure shows the elevator maintenance status; Design 10, reference figure shows the earthquake early warning status.

Techniques for improving how LSR is performed are disclosed. A service accesses a depth image, a pose correction matrix, and a color image. The service extracts a carrier geometry from the depth image. The service forward projects the LSR carrier geometry by multiplying each vertex of the LSR carrier geometry with the pose correction matrix. While the GPU is operating in a shadow map mode, the service causes the GPU to perform a rasterization process to produce a UV map and a Z buffer. The service discards the UV map, resulting in the per pixel UV corrections included in the UV map also being discarded. The service recovers the per pixel UV corrections using the Z buffer. The service uses the recovered per pixel UV corrections to resample the color image, resulting in generation of a corrected color image.

The invention relates to a dynamic vision SLAM (Simultaneous Localization and Mapping) optimization method and device. A bounding box detection result is divided into a prior dynamic object detection box and a static object detection box according to semantic information in an RGB (Red, Green and Blue) image; motion consistency analysis is carried out on the prior dynamic object detection frame through an optical flow method, a first type of the prior dynamic object detection frame is determined, and the first type comprises a real dynamic object class and a static prior dynamic object class; determining a second type of a feature point according to the depth value of the feature point in the prior dynamic object detection frame of each real dynamic object class in the depth image, wherein the second type comprises a foreground point and a background point; calculating an initial pose result; and optimizing the initial pose result to obtain a final pose result. According to the method, the distribution characteristics of reprojection errors are analyzed in real time, and robust kernel function parameters are dynamically adjusted, so that the adaptability of an algorithm to a detection failure scene is enhanced, interference of non-Gaussian noise is effectively suppressed, and the positioning precision and robustness in a dynamic environment are improved.

An exemplary device, according to the present disclosure, provides a housing, an AR lens, a projector, a memory, and a control system. The control system and memory determine when the user is confused. The control system then provides for projecting a reorientation graphic via the projector directed at the AR lens. In some implementations, the reorientation graphic provides real time information and reminders to a user as they proceed through life. The exemplary device enables a user to maintain greater independence and a more normal lifestyle with the assistance of the reorientation graphic.

The invention relates to the technical field of image processing, and discloses a shadow rendering method and device, equipment and a storage medium. The shadow rendering method comprises the following steps: projecting the contour of a foreground image matting to a virtual ground in a preset virtual three-dimensional scene to generate a shadow projection; comparing visual depths of overlapped pixels in the shadow projection and the foreground matting to adjust a visible range of the shadow projection; and carrying out color mixing on the shadow projection in the visible range and the overlapped pixels of the virtual ground so as to render the shadow projection in the visible range to a preset virtual three-dimensional scene. By means of the method, the shadow effect of foreground image matting can be rendered in the virtual three-dimensional scene, and the sense of reality of shadow rendering and the real-time rendering efficiency are improved.

A visual effects system (12) includes a mirror structure (18) including a mirrored surface (32) and a non-mirrored surface (34). The mirrored surface (32) extends about the non-mirrored surface (34) and the mirrored surface (32) and the non-mirrored surface (34) face a guest area (14). The visual effects system (12) also includes one or more props (20) positioned proximate to the mirrored surface (32) and extending between the mirror structure (18) and the guest area (14), wherein the one or more props (20) are oriented such that the one or more props (20) is viewable from the guest area (14) and a reflection of the one or more props (20) provided by the mirrored surface (32) is viewable from the guest area (14), and a projector (24) oriented to project one or more images onto the non-mirrored surface (34).

The invention relates to the technical field of cartoon development graphic processing, and particularly discloses an operation method for cartoon development, which comprises the following steps: establishing a cartoon style shadow sampling coordinate system, configuring stylized shadow parameters based on a virtual light source visual angle, and generating a multi-frame sampling data sequence through frame-by-frame spatial offset; carrying out depth comparison on the sampling data and scene geometric data, and generating primary shadow data with stylized noise characteristics in combination with animation rendering characteristics; generating a stylized credibility distribution diagram through motion compensation reprojection and artistic feature consistency comparison; establishing a mapping relation between visual features and fusion weights based on the distribution map, and executing art-oriented adaptive filtering processing to generate high-quality shadow data; and finally, integrating shadow data into a rendering pipeline, and dynamically optimizing a subsequent sampling strategy by analyzing visual continuity.

The application provides a model rendering method and system based on mixed reality interactive creation, and relates to augmented reality technology, comprising: calculating an optimal rendering view angle based on the geometric structure of a three-dimensional model; obtaining a preset optimal viewing angle, and calculating a final rendering view angle by weighting the optimal viewing angle and the optimal rendering view angle; projecting the three-dimensional model based on the final rendering view angle to generate a baseline sketch map; obtaining an interactive coloring image with the baseline sketch map as a bottom map; identifying the interactive coloring image, extracting a regional image matched with the baseline sketch map, and performing geometric correction; extracting color features from the regional image, coloring the baseline sketch map based on the color features, and generating a color texture map; and rendering the three-dimensional model based on the color texture map. The application enables users to input creativity in a hand-drawing manner, automatically generates a high-quality map conforming to a target three-dimensional object, and finally renders and displays in real time in an augmented reality environment.

A method is provided for manufacturing a security document by intaglio-printing a rendering of a three-dimensional object on the security document. The method comprises at least the following: S1) providing a model (1) of the object with a plurality of surfaces (A, B, C, D) in three-dimensional space; S2) attributing, to each point on the surfaces (A, B, C, D), a grayscale or color value; S3) using mapping functions (M) for mapping the surfaces (A, B, C, D) into two-dimensional parameter spaces, thereby defining grayscale or color images (2) in the parameter spaces, wherein the grayscale or color value of each point in said images is given by the grayscale or color of the point on the surface corresponding to the point of the image; S4) rendering, in the parameter spaces, the grayscale or color images (2) as halftone images (3) using halftone patterns, wherein at least some of the halftone patterns are different from each other; S5) using inverses of the mapping functions for mapping the halftone images (3) as textures onto the surfaces (A, B, C, D); S6) projecting the model with the surfaces (A, B, C, D) carrying the textures into a two-dimensional printing plane using a three-dimensional projection (P), thereby projecting said halftone images (3) and textures into the printing plane; S7) rendering, within the printing plane, the surfaces (A, B, C, D) with the textures into a two-dimensional array of printing pixels and forming an intaglio printing plate (5) with recesses corresponding to the array of the printing pixels; and S8) printing, using the intaglio printing plate (5), the rendering of the three-dimensional object onto said security document.

A computer-implemented method for transforming speech into visual text, including obtaining a speech signal encoding a text of a speech; selecting a portion of the text at least based on a projection mode of a writing instrument; outputting the portion of the text towards a projector configured to project the portion of the text to a user of the writing instrument. The writing instrument may be used by the user to take notes of and during the speech and may include the projector, wherein the projector may be configured to project the portion of the text onto a writing surface being used by the user to write on with the writing instrument. The method may include projecting the portion of the text to the user.

Techniques for improving how LSR is performed are disclosed. A service accesses a depth image, a pose correction matrix, and a color image. The service extracts a carrier geometry from the depth image. The service forward projects the LSR carrier geometry by multiplying each vertex of the LSR carrier geometry with the pose correction matrix. While the GPU is operating in a shadow map mode, the service causes the GPU to perform a rasterization process to produce a UV map and a Z buffer. The service discards the UV map, resulting in the per pixel UV corrections included in the UV map also being discarded. The service recovers the per pixel UV corrections using the Z buffer. The service uses the recovered per pixel UV corrections to resample the color image, resulting in generation of a corrected color image.

A shadow rendering method is performed by an electronic device. The method includes: determining scene ranges corresponding to a plurality of first levels, each scene range being a first visible scene range of a virtual scene corresponding to a respective one of the plurality of first levels; for a respective one of the plurality of first levels: determining a first virtual item whose shadow is present in a first visible scene range corresponding to the first level; determining a first projection depth map corresponding to the first level based on the first virtual item, the first projection depth map being configured for describing distances between a light source and a plurality of first vertices on a surface of the first virtual item facing the light source; and rendering the shadow of the first virtual item in the first visible scene range based on the first projection depth map.

The application discloses a kind of autoregressive scene layout generation methods based on three-dimensional diffusion model, comprising: using pre-trained basic three-dimensional diffusion model, three-dimensional grid is encoded into voxel sparse latent variable by variational autoencoder;Using the frozen large language model as text encoder extracts the conditional vector of scene layout generation task text, and infers the placement order of target object;From empty scene, sequentially perform the following autoregressive layout operation on each target object, obtain the estimated pose of each target object;All target objects are placed in the scene coordinate system according to the estimated pose, and the final three-dimensional layout is obtained.The application directly generates layout in native three-dimensional space, fundamentally eliminates the spatial modeling error caused by two-dimensional projection, and greatly improves the physical rationality of layout.

This invention relates to a graphics API recommendation method based on a multi-scene lightweight adapter, comprising the following steps: acquiring a public image dataset D, and constructing a recommendation model M based on a Transformer decoder; selecting any image data from D for data preprocessing to obtain standardized image data, and then performing image block segmentation and feature embedding operations sequentially to obtain an initial visual feature sequence; inputting the input into the first Transformer decoder to obtain the core branch backbone features at layer l; inputting the input into MLP branches and branches respectively to obtain the self-attention feature matrix and the preliminary dimensionality reduction projection feature matrix; inputting the input into the output to obtain a low-dimensional matrix, and inputting the input to obtain a high-dimensional matrix, calculating the final feature matrix after passing through layers, and sequentially passing it through all subsequent Transformer decoders to obtain a graphics API probability sequence; then training the recommendation model, and obtaining the trained recommendation model M' after completion.

The invention relates to the technical field of graphical user interface generation and dynamic layout control, and discloses an airborne human-computer interface dynamic generation method for uncertain number of graphs. Establishing a rectangular coordinate system at the center of the interface, obtaining the maximum layout radius of width and height and the radius of a circular element, and forming feasible point judgment; constructing a logarithmic spiral trajectory and an angular step length, and deducing a growth coefficient; generating a polar angle according to steps, calculating a polar radius, and crossing a boundary, namely a cutoff displayable quantity; converting polar coordinates into a right-angle candidate set; verifying according to a center distance, performing single displacement and border crossing projection on a post-added point, deleting non-compliance and performing synchronous binding; collecting state values, normalizing and sorting, generating a new angle, a new radius and a new coordinate, re-numbering after re-checking, and synchronizing priorities; drawing a connecting line for the associated primitives according to a threshold value; and finally outputting a coordinate, a priority matrix and a connection set.

The invention discloses an OCR text sorting and grouping method and system and a storage medium, and the method comprises the steps: under a PP-OCRv3 framework, generating a text detection box through employing a differentiable binary text detection algorithm, and carrying out the recognition of a text content through an SVTR model; in the sorting stage, primary sorting is carried out according to Y-axis coordinates, different rows are segmented according to a 200-pixel row spacing threshold value, sorting is carried out in the rows according to X-axis coordinates, and a 300-pixel horizontal spacing threshold value is set to limit a reading sequence; the inter-row Y-axis tolerance, the horizontal spacing threshold value and the Y-axis strict tolerance are set by combining the image height proportion, and reasonable merging of the slightly biased text box is achieved; on the basis of local density optimization of Y-axis projection distribution, text detection boxes with similar projection distribution are gathered into the same row, and sorting deviation caused by inclination or deformation is corrected by using X-axis local density, so that a text sequence conforming to a natural reading sequence is output, and output logicality and readability are improved in a complex typesetting and handwritten scene.

This invention relates to the field of image processing technology, specifically disclosing a zero-shot text attribute graph learning method based on prototype mutual prompting enhancement. The method calculates instance-aware contrast loss and constrains the similarity of the same node's embedding in two views. It performs PCA dimensionality reduction on the token embedding of the LLM, extracts the first P principal components to form an aligned coordinate system C, maps the components to this coordinate system, and calculates the informative-aware contrast loss. It freezes the pre-trained GNN and the LLM backbone network, extracts node structural embeddings Z from the pre-trained GNN, generates K prototype embeddings through K linear projectors, and injects them as soft prompts into the LLM instructions. The LLM outputs K expert prompts. It calculates the weights of each expert prompt through a router model. It freezes the LLM training of related projectors and routers, freezes the GNN training of the LLM prompt adaptation layer, and iterates and optimizes 1-3 times. This method can cover complex cross-modal information in text and images while reducing the accumulation of zero-shot transfer bias.

1. The name of the design product: toy (projection lantern). 2. The use of the design product: the design product is used for children to play. 3. The design points of the design product: in shape. 4. The picture or photo that best indicates the design points: perspective drawing.

The invention discloses a zero-sample stylized three-dimensional graphic content generation method and system based on three-dimensional Gaussian sputtering. The method comprises the following steps: down-sampling an input three-dimensional Gaussian sputtering scene through an encoder in an auto-encoder and extracting three-dimensional scene features; after the three-dimensional scene features are projected to a feature space aligned with the two-dimensional style image features through a multi-layer perceptron, style features from the two-dimensional style image are injected into the three-dimensional scene features through adaptive instance normalization, and features after stylization conversion are obtained; and performing up-sampling on the stylized and converted features through a decoder in the auto-encoder, and reconstructing a stylized and converted three-dimensional Gaussian sputtering scene. According to the method, real-time and high-quality style migration of any three-dimensional scene and any style image can be realized, time-consuming optimization for each new scene is not needed, and the efficiency and the flexibility of three-dimensional content creation are greatly improved.

This invention discloses a multimedia light painting collaborative interactive system integrating light and shadow art and intelligent sensing, comprising an interaction perception module, an intent understanding and mapping module, a collaborative interaction management module, an intelligent graphics rendering and output module, and an environment adaptation module. Through multimodal sensor fusion, the system achieves high-precision motion capture and complete, continuous image acquisition under complex lighting conditions. It can intelligently analyze user motion characteristics to understand creative intent and utilize GPU parallel computing and AI style transfer to generate personalized light and shadow effects in real time. The system supports multi-user collaborative interaction and can automatically trigger collective creation events. The environment adaptation module can dynamically adjust the projection output to ensure optimal visual effects across the entire scene. This invention represents a leap from static recording to dynamic intelligent co-creation, providing an immersive and highly interactive new experience.

The application provides a projection method and device based on cocos2d-x, which comprises the following steps: superimposing textures of all projection nodes and then rendering to obtain a sprite; and projecting and displaying according to the sprite. Compared with the prior art, the prior art directly superimposes the projection nodes after rendering each node separately and then changes the transparency, thereby causing superimposition problems. By using the method of superimposing the textures of all projection nodes and then rendering to obtain a sprite, the consistency of color values and transparency is ensured, thereby avoiding superimposition problems and color deepening. Furthermore, the textures of all projection nodes are obtained by using RenderTexture and rendering, the textures of the projection nodes are drawn only by obtaining the projection nodes, and projection is performed according to the sprite, so that the creation and destruction of the projection nodes are not required, and the performance overhead is reduced.

1. The name of the design product: the support of the projector. 2. The use of the design product: the whole is used for projection, and the part is used for supporting the projector. 3. The design points of the design product: the shape of the solid line part. 4. The picture or photo that best shows the design points: perspective view 1. 5. Other circumstances that need to be explained: the solid line in the view defines the part claimed, and the dashed line defines the part not claimed.

The invention relates to the technical field of image data processing, and discloses a Chinese ancient book text restoration method and system based on big data, and the method comprises the steps: constructing a homologous font skeleton index database, and extracting the topological nodes and curvature features of the same version stroke skeleton; recognizing a damaged area and extracting a residual skeleton, and matching an optimal homologous skeleton; constructing a thin-plate spline interpolation function and executing non-rigid geometric deformation to generate a target guide skeleton; calculating a damaged region neighborhood background image gradient field to construct a background interference tensor field, and extracting an interference direction vector; constructing an interference vector orthogonal complement space and projecting the skeleton direction vector to the interference vector orthogonal complement space to obtain a net skeleton direction vector; a structure tensor field is reconstructed based on a net skeleton direction vector, texture mapping is guided, through topological isomorphic mapping and orthogonal tensor field constraint, the coupling problem of stroke structure distortion and background noise interference in ancient book repair is solved, and high-fidelity structure reconstruction and texture fusion under a complex engraving style are achieved.