66 results about "Morphing" patented technology

This invention relates to the fields of computer vision and computer graphics, and provides a method and system for dynamic human body reconstruction from monocular video based on 3D Gaussian splashing. The method includes the following steps: data preprocessing; initialization of the normalized space 3D Gaussian; deformation of the normalized space 3D Gaussian to an intermediate pose space to obtain a non-rigidly deformable 3D Gaussian and pose-related features; transformation of the non-rigidly deformable 3D Gaussian to the observation space using linear blending skinning to obtain the observation space 3D Gaussian; decoding the viewpoint-related color based on Gaussian color features, pose-related features, and viewpoint direction; constructing a total loss function including a normal consistency regularization term to optimize the 3D Gaussian attributes and network parameters; and rendering the target human body image using a differentiable Gaussian splash rasterizer. This invention enables rapid and fully automatic reconstruction from monocular video to a high-fidelity, animable human body model, applicable to fields such as virtual reality and film production.

The application relates to a deformation field prediction method based on image quality evaluation and adaptive pre-training. Wavelet packet decomposition is performed on a reference image and a deformation image of a collected component to obtain coefficients of each subband of each layer of the images, and Pearson correlation coefficients of each subband are calculated based on all the coefficients corresponding to the subband. The subbands are screened based on L2 norms of coefficient vectors of the subbands and the Pearson correlation coefficients, and the screened coefficient vectors of the subbands are reconstructed to obtain reconstructed reference images and reconstructed deformation images. Quality evaluation indexes of each reconstructed sample pair are calculated. Training branches to which the corresponding reconstructed sample pairs belong are determined based on the quality evaluation indexes, a U-Net model is preliminarily trained by using a preset supervised training strategy based on the reconstructed sample pairs of different training branches, the U-Net model is unsupervisedly trained, and displacement fields and strain fields are predicted based on the trained U-Net model.

This application provides a method, system, medium, and device for human-centered video scene reconstruction and separation. The method includes: for a first-person perspective video sequence, initializing a 3D Gaussian set covering the background, hands, and objects based on a priori knowledge of structure recovery from motion; assigning a learnable dynamic category probability vector to each Gaussian point in the 3D Gaussian set; constructing a dedicated deformation branch; according to the learnable dynamic category probability vector of each Gaussian point, assigning each Gaussian point to the dedicated deformation branch for processing through a preset soft-hard two-stage routing mechanism, determining the Gaussian points processed by the dedicated deformation branch; rendering the Gaussian points processed by the dedicated deformation branch to determine the 4D scene reconstruction image and the decomposed reconstruction images of the background, hands, and objects. This application achieves 4D scene reconstruction of human-centered video and explicit, fine-grained separation of the background, hands, and objects.

The application discloses a kind of based on parameterization shape's deformation character dynamic typesetting rendering method and system, the method includes: according to the preset intensity parameter value range of character rendering, respectively generate corresponding Bezier curve discrete sampling dataset;The curve sampling data corresponding to each intensity parameter is converted into rgb color value, and is stored in texture carrier or using texture array according to line;Synchronous generation auxiliary array, store the maximum vertical coordinate value in the curve sampling data corresponding to each intensity value;The horizontal typesetting processing of to-be-rendered character is carried out, generates temporary image and records original width and height;According to the selected intensity parameter, obtain the corresponding maximum vertical coordinate value in auxiliary array, calculate target image height;Temporary picture, intensity parameter, original width, target image height and rgb color value are input into shader program, and are mapped to Bezier curve deformation space, generate final rendering image.The application can realize efficient, diversified character deformation rendering.

The present invention sets forth techniques for predicting motion in a 3D model. The disclosed techniques include receiving one or more Gaussian primitives representing a 3D scene and receiving one or more control inputs describing one or more primary motions associated with one or more objects. The techniques also include generating, via a first trained machine learning model, a dynamic state based on the one or more control inputs, and generating, via a second trained machine learning model, one or more deformed Gaussian primitives based on the dynamic state and the one or more Gaussian primitives. The techniques further include generating a 2D representation of the 3D scene, and generating an output sequence based on the 2D representation, wherein the output sequence depicts both the primary motions associated with the one or more objects and one or more secondary motions associated with the one or more objects.

The application relates to a WebGL-based three-dimensional local terrain protrusion weakening method, and belongs to the technical field of computer graphics visualization, which comprises terrain grid loading, loading terrain data in the form of a grid into a scene in a WebGL rendering engine, storing information in the form of a.terrain format binary file, then performing grid analysis, loading block terrain data according to a terrain level of detail strategy, analyzing terrain data, loading and analyzing downloaded binary file data according to terrain grid levels and row and column numbers, compressing the data by using a zig-zag coding method, then performing vertex height dynamic adjustment, and finally performing terrain triangular net reconstruction. The WebGL-based three-dimensional local terrain protrusion weakening method can weaken terrain protrusion problems by modifying the vertex elevation of a terrain data-free area, does not cause distortion of local high-precision terrain and models in a scene, can weaken terrain cliff problems existing in three-dimensional local terrain protrusions, and does not deform the terrain and models in the scene.

Approaches are disclosed for enhancing frame rate and visual smoothness in real-time video streams through multi-frame interpolation. A classification neural network analyzes two sequential frames, outputting confidence scores that indicate the reliability of motion data for each pixel. These scores determine whether a pixel's motion is accurately described by motion vectors or should be treated as static. The classification results are reused to generate intermediate frames by warping the original frames based on the motion characteristics. Blending weights are calculated by combining warped motion vector confidence values with static values, and a second neural network refines the alignment and blending of candidate frames. This second network predicts intermediate flows and generates new blending weights, which are used to warp and blend the candidate frames, ultimately producing a final interpolated frame that enhances visual smoothness and consistency in the video stream.

A processing system is a processing system configured to communicate with a server configured to store characteristic data indicating characteristics of a medium, the processing system including: a feeder configured to feed the medium from a roll; a printer configured to print an image on the medium fed from the feeder; a dryer configured to dry the medium on which the image is printed by the printer; a winder configured to wind the medium dried by the dryer; an imager configured to capture an image of the medium; and a controller configured to acquire a captured image from the imager, and the controller is configured to transmit type data indicating a type of the medium to the server, acquire the characteristic data corresponding to the type indicated by the type data from the server, determine based on the captured image whether the medium is deformed, and control a processing parameter based on the characteristic data when the controller determines that the medium is deformed.

This application relates to a high-fidelity real-time rendering method, apparatus, device, and storage medium for dynamic digital humans. It includes receiving multimodal driving signals, mapping them to expression and pose parameters, inputting a parameterized human model to generate standard spatial geometry and skinning weights, and constructing a three-dimensional Gaussian sputtering field to deform to the target pose to obtain dynamic geometric data; arranging sparse virtual cameras in the target pose space, and simultaneously rendering and generating a sparse reference view set containing color and depth within a single rendering call; determining dense virtual viewpoints based on the display device's viewpoint parameters, and generating depth range textures for each dense viewpoint through downsampling depth reprojection and point sputtering techniques; determining ray step intervals using the depth range textures, sampling along the rays and fusing signed distance values ​​truncated from multiple viewpoints to generate surface position textures; projecting the surface positions onto sparse view sampled colors, fusing to generate a dense view color image, and adapting it for output to different display devices.

Systems and methods are disclosed for training and using a digital human foundational model (DHFM) comprising a generative adversarial network (GAN) generator. For instance, the method may include obtaining one or more inputs comprising pose information indicating a three-dimensional (3D) pose representation of a human and processing the one or more inputs using a mapping network to generate intermediate latent code. The method may further include processing the intermediate latent code using the trained generator to generate texel-aligned Gaussian maps that align Gaussian attributes to a coarse mesh template of the human and performing linear blend skinning and deformation on the texel-aligned Gaussian maps to obtain modified texel-aligned Gaussian maps. The method may also include processing the modified texel-aligned Gaussian maps using a multi-part renderer to generate a synthetic human representation of the human indicating facial and hand features of the human.

This invention relates to the fields of computer vision, text processing, font generation, and adaptive learning, and specifically to an adaptive text display method and system for assisting memory. The method involves inputting text content to be memorized, which can be a text string and its corresponding vector font file, or a rasterized image containing the text content. Based on the input type, auxiliary display elements are generated or acquired to partially replace or cover the original text's visual form. Based on the user's memory proficiency index and / or the semantic features of the text content, several target layout rules are selected from a predefined layout rule library. According to the target layout rules, the original text data and the auxiliary display elements are logically fused to output the desired adaptive memory-assisting display interface. This invention elevates memory-assisting technology from a simple font transformation to a context-aware intelligent interactive interface, breaking the limitations of traditional font technology application scenarios.

A computer-implemented method and system refer to an image encoder that receives a digital image as input. The image encoder generates a weighting map using a preceding feature map. The preceding feature map is generated using pixels of the digital image. The weighting map is generated based on Lie data associated with the digital image. A homographic transformation is interpolated between two planar projections of the digital image using at least the weighting map and a homography matrix. The homography matrix provides a mapping between the two planar projections of the digital image. Homographically transformed kernels are generated by applying the homographic transformation to convolution kernels.The homographically transformed kernels are applied to the preceding feature map to perform folding on different planar areas appearing in the digital image and to generate a new feature map that is used for a computer vision task involving three-dimensional (3D) perception.

A method includes obtaining, using at least one processing device of an electronic device, multiple sets of training image frames, where each set of training image frames has an associated ground truth image. The method also includes applying, using the at least one processing device, motion blur and warping to the multiple sets of training image frames in order to generate additional sets of training image frames. In addition, the method includes training, using the at least one processing device, a machine learning model to align image frames and remove motion blur from the image frames based on at least the additional sets of training image frames and the ground truth images.

The present application relates to the technical field of computer vision, in particular to a kind of automatic matching method of picture template and material, comprising the following steps: S1: matching, in the container in template unit, select a suitable material picture for each container;S2: positioning and cutting, the matched picture is cut to the size suitable for filling template.The present application analyzes the container demand through language big model, score mechanism dynamically balances picture type, main body, interaction, adapts to different template demand, and uses multi-modal model to detect material picture main body, expands boundary to maximize to retain background, according to template size proportion, central cutting, avoids deformation, ensures visual aesthetic appearance.At the same time, from matching to cutting whole process does not need manual intervention, supports API access e-commerce system, efficiency is improved by more than 90%, adapts to the template rule of multiple overseas e-commerce platform, reduces cross-platform operation cost.

The present disclosure belongs to the field of artificial intelligence, and proposes a scene text image generation method, system, device and storage medium, comprising: obtaining input prompt text for scene text image generation, and generating glyph images and initial background images based on the input prompt text respectively; positioning the text position in the initial background image to generate a mask image; introducing a scene image, and transforming the glyph image based on the scene image, the initial background image and the mask image to obtain a deformed text image, wherein the scene image is determined according to the initial background image and the mask image; generating an auxiliary image according to the deformed text image and the scene image, and generating a scene text image based on the auxiliary image. The proposed glyph separation effectively improves the accuracy of generated text; the proposed scene perception improves the realism of generated text; in view of the problem of visual quality decline caused by directly fusing glyph and background images, the proposed text repair realizes multi-language scene text image generation.

1. Name of the product in this design: Toy (Airplane Transforming Robot). 2. Purpose of this design: This design is intended for children to play with. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: 3D rendering 1.

The application discloses a morphing text detection method and device and a storage medium. The method comprises the following steps: extracting image features of a to-be-processed image, the to-be-processed image being an image collected from a specified text carrier, and the specified text carrier being easy to deform under stress; inputting the image features into a control point detection network to obtain an edge control point sequence; inputting the edge control point sequence into a differentiable B-spline layer to perform matrix multiplication calculation on a base function matrix and the edge control point sequence, thereby obtaining a B-spline curve point sequence, and the base function matrix being a two-dimensional matrix composed of B-spline base function values; and generating a text detection frame based on the B-spline curve point sequence. The operation of the differentiable B-spline layer is differentiable in a deep learning framework, and the gradient can be back propagated to the parameters of the control point detection network through the differentiable B-spline layer, thereby realizing integrated end-to-end optimization of text detection and curve fitting and significantly improving the performance of the model in a complex morphing text detection task.

This invention discloses a method for detecting and recognizing tilted and distorted text in natural scenes, comprising the following steps: preprocessing a natural scene image containing tilted, curved, or perspective-distorted text, and obtaining the stroke width information of each pixel in the image and the corresponding edge pixel relationship through edge detection and stroke width calculation. Based on the consistency of stroke width and the edge correspondence, connected region aggregation is performed on the pixels in the image to obtain candidate text regions, and a text stroke structure is constructed. The stroke center lines are further extracted and control points are generated. The weights of the control points are determined based on the stroke structure information, and a weighted control point set is constructed. Based on this control point set and candidate regions, a deformation mapping is generated using the moving least squares method to correct the tilted and distorted text. The corrected text regions are then filtered and character recognition is performed, outputting the text detection and recognition results. This invention is applicable to text processing in complex backgrounds.

This application provides a display method, apparatus, computer device, computer-readable storage medium, and computer program product. The method includes: acquiring first reference information for each deformable component of a virtual object in a preset posture; during the movement of the virtual object, acquiring a first starting position and a first ending position for each deformable component; determining a first deformation parameter for the deformable component based on the first reference information, the first starting position, and the first ending position; updating the motion structure model of the virtual object using each of the first deformation parameters to obtain a deformed motion structure model; and displaying the virtual object based on the deformed motion structure model. This application improves the realism of virtual object display.

The application discloses a semantic segmentation enhancement method for automatic loading and unloading of bagged materials, relates to the technical field of automatic loading and unloading, and comprises the following steps: replacing an image encoder of a SAM3 model with a deformable attention module to obtain optimized local fine-grained feature extraction; adding an edge perception auxiliary decoding head in a mask decoder; setting a scene-adaptive Prompt encoding strategy in the model; and reconstructing through an edge loss function to obtain a semantic segmentation model for the automatic loading and unloading of bagged materials. The application greatly reduces the calculation complexity through the deformable attention module, reduces the single-frame inference time of the model, and meets the real-time segmentation demand of the industrial full-automatic loading and unloading scene. The zero-shot detection advantage is retained, and the text Prompt template can be modified to adapt. The scene-adaptive composite Prompt encoding and inter-frame propagation strategy realize batch continuous automatic segmentation of bagged materials.

The present invention can be applied to the technical field of computer vision and graphics. Provided are a method for disentangled reconstruction of a dynamic digital human, and an electronic device and a storage medium. The method comprises: using a technique for reconstructing three-dimensional shapes of a human body and garments from a monocular human body video, and using a representation method in which explicit geometry is combined with an implicit signed distance field (hmSDF), such that high-quality reconstruction of a dynamic disentangled digital human from a monocular video can be achieved. In the method, garments and a human body are separated and separately modeled, optimized hmSDF is used to achieve accurate segmentation of visible regions, and an SMPL model is also used to complete occluded human body regions, thereby ensuring the consistency and fidelity of the overall geometry. A linear blend skinning (LBS) deformation field and a non-rigid deformation field are used to capture human body motion and detailed variations, such that high-fidelity and continuous spatio-temporally disentangled human body and garment geometry is ultimately generated.

To provide a device for measuring the size of an object that can measure the appropriate size of an irregularly shaped object. [Solution] The object size measuring device (10) is equipped with depth sensors (14V, 14H) that acquire a 3D image of the object to be measured. A preprocessing means (22, 40e) preprocesses the 3D image to remove noise and other imperfections to generate a preprocessed 3D image. This preprocessed 3D image is converted into a voxel assembly by a voxelization means (22, 40f). A deformation means (22, 40g) deforms the voxel assembly, for example, to eliminate irregularities in the voxel assembly, thereby generating a deformed voxel assembly. A rotation means (22, 40h) rotates the circumscribing rectangle of the projection image on at least one slice plane of the deformed voxel assembly to find the rotation angle at which the four sides of the circumscribing rectangle are shortest, and a calculation means (22, 40i) calculates the length, width, and height of the deformed voxel assembly at that rotation angle. [Effect] Allows for efficient measurement of the correct size of irregularly shaped objects.

This disclosure relates to the field of computer graphics technology, and provides a method, apparatus, device, medium, and program product for clothing adaptation of virtual characters. The method for clothing adaptation of virtual characters includes: first, establishing a three-dimensional spatial mesh containing a source virtual character model and a target virtual character model with the same topological structure, as well as a source clothing model; then, based on the positional correspondence between the source and target virtual character models, and combined with smoothing constraints to limit the difference in deformation vectors between adjacent mesh points, calculating the deformation vector of each mesh point in the three-dimensional spatial mesh; finally, using the deformation vector of each mesh point, mapping and transforming the source position of each vertex in the source clothing model to the target position, thereby generating a target clothing model adapted to the target virtual character model. This disclosure can effectively maintain the geometric characteristics of the clothing model, reduce distortion during the adaptation process, and improve visual effects.

This invention discloses an end-to-end video stabilization method using a decoupled spatiotemporal Transformer, belonging to the field of computer vision. The method constructs a multimodal input by fusing RGB images and bidirectional optical flow, and directly regresses a pixel-level dense deformation field via a spatiotemporal Transformer-UNet network to achieve image stabilization. Its core lies in the decoupled spatiotemporal block embedded within the network, which alternately executes spatial self-attention and trajectory-aware temporal interactive attention based on a Lagrange perspective. The latter utilizes optical flow priors to distort and align features of adjacent frames along the motion trajectory, fundamentally solving the feature misalignment problem caused by severe camera shake. Furthermore, by designing an unsupervised training framework consisting of motion smoothing, photometric reconstruction, and geometric consistency loss, camera shake and foreground motion can be implicitly separated. This invention exhibits excellent performance in terms of cropping rate, distortion rate, and stability metrics, demonstrating strong robustness.