Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

460 results about "Single view" patented technology

Structuring and rich debugging of inputs and outputs to large language models

The disclosure is directed to methods and systems for improving interactions with a Large Language Model (LLM). An artificial intelligence system (AIS) can receive user inputs via a graphical user interface indicating a task to be performed by the LLM, one or more tools which may be accessed by the AIS in response to tool calls from the LLM, and an output schema for structuring a format of a response from the LLM. The AIS can generate a prompt for the LLM based on the user input. The prompt can include indications of the one or more tools, one or more example tool operations, the task to be performed, and an indication of the output schema. The AIS can include a debugging application or module enabling rich debugging of language model interactions in a single view.
Owner:PALANTIR TECHNOLOGIES INC

Tumor area double-view detection method and system based on image modality and view

The invention provides a tumor area double-view detection method and system based on image modalities and views, and relates to the field of medical image analys.The method comprises the steps that multi-modal image data are collected from the body of a patient, and image data of different view angles at the same position are obtained according to time frames; extracting multi-modal features from the image data of different modalities and different view angles, and performing weighted fusion on the extracted multi-modal features to form a fused feature map; inputting the fused feature map to train and predict the 3D-CNN network so as to identify and position a tumor region, and marking the tumor region by using a mask output by the segmentation model; calculating the pixel number of the tumor area in each layer of image according to the marked mask, and multiplying the pixel number by the volume of each pixel to obtain the volume of the tumor area; and the segmentation results of each layer of image are superposed, and a three-dimensional model of a tumor region is generated by using 3D reconstruction, so that the limitation of single-mode and single-view image data in tumor detection is overcome.
Owner:SICHUAN CANCER HOSPITAL

Neural head avatar construction from an image

Systems and methods are disclosed that animate a source portrait image with motion (i.e., pose and expression) from a target image. In contrast to conventional systems, given an unseen single-view portrait image, an implicit three-dimensional (3D) head avatar is constructed that not only captures photo-realistic details within and beyond the face region, but also is readily available for animation without requiring further optimization during inference. In an embodiment, three processing branches of a system produce three tri-planes representing coarse 3D geometry for the head avatar, detailed appearance of a source image, as well as the expression of a target image. By applying volumetric rendering to a combination of the three tri-planes, an image of the desired identity, expression and pose is generated.
Owner:NVIDIA CORP

Satellite three-dimensional reconstruction method in complex illumination environment

The invention relates to the technical field of three-dimensional modeling, and provides a satellite three-dimensional reconstruction method in a complex illumination environment, which comprises the following steps: acquiring images under different illumination conditions, and extracting image features by combining multi-view collaborative observation; performing feature point and multi-view data registration, generating a three-dimensional model and analyzing illumination change; performing system initialization by using the initially matched feature points, and establishing an initial three-dimensional point cloud model; sparse point cloud is recovered through multi-view stereoscopic vision, and motion recovery is carried out; a 2D diffusion model is used to complement the shielded area, and laser radar data is combined to carry out densification processing on the sparse point cloud; a three-dimensional grid model is generated, and texture mapping is carried out by a multi-view image; light compensation is adopted, a polarized light separation technology is used, and ambient light and specular reflection light are separated; according to the method, through multi-view-angle collaborative observation, the shielding problem under a single view angle is reduced, the 2D diffusion model is used for complementing the shielded area, and the completeness and precision of three-dimensional reconstruction are ensured.
Owner:未来宇航(上海)航天科技有限公司

Deep dense document recall method based on multi-view vector fusion

The invention relates to the technical field of natural language processing and information retrieval, and discloses a deep dense document recall method based on multi-view vector fusion, comprising the following steps: S1, preprocessing user query and candidate documents to generate a standardized text; s2, respectively constructing multi-view semantic vector representation of the query and the document, wherein multiple views at least comprise a keyword view, a semantic extension view and an intention view; s3, calculating a similarity score between the query and the document for each semantic view, and fusing the scores of all the views through a dynamic weight; and S4, sorting the documents according to the fusion score, and returning a front Top-K result. By constructing multi-dimensional semantic representation of keywords, semantic extension and an intention perspective, multiple semantic information such as term accurate matching, context association and task target consistency is effectively fused. The dynamic weight distribution mechanism adaptively adjusts the contribution degree of each view angle according to the query content, and overcomes the defect of insufficient semantic coverage of a single view angle.
Owner:NANCHANG HANGKONG UNIVERSITY

Pig single-view three-dimensional reconstruction method based on deep learning

The invention discloses a pig single-view three-dimensional reconstruction method based on deep learning, and the method is called Magfup. In order to solve the problems that single-view three-dimensional reconstruction viewpoint learning is unstable and training is difficult due to lack of real three-dimensional data, an implicit-explicit combined three-dimensional representation method is adopted, high-dimensional data features subjected to super-resolution optimization are combined, a high-density pig three-dimensional grid can be reconstructed only through a single image, and the problem that training is difficult is solved. And data acquisition cost and equipment requirements are reduced. According to the method, the three-dimensional grid surface is optimized through the Laplacian smoothing method, and the reconstruction quality is improved. According to the invention, the model can learn the general shape and posture characteristics of the pig, predicts the three-dimensional posture, shape and texture information of the specific pig, and more accurately and comprehensively reflects the behavior characteristics of the pig. Therefore, the method can adapt to complex and changeable environments and pig behaviors in a pig farm, and a more comprehensive and accurate data basis is provided for pig behavior analysis.
Owner:SOUTH CHINA AGRICULTURAL UNIVERSITY

Single-view three-dimensional modeling method and system based on diffusion model

The invention discloses a single-view three-dimensional modeling method and system based on a diffusion model. The method comprises the steps of firstly obtaining a single image as input; the received images are input into a multi-view diffusion model, then the generated multi-view images are input into a ViT encoder, and each view image is converted into an image token through the ViT encoder; inputting the image token into a three-plane decoder to generate a three-plane representation; finally, contour surface extraction is carried out on the three-plane representation by using FlexiCubes, so that the three-plane representation is converted into grid representation; and 3D grid rendering: rendering the generated grid representation to generate a 3D model view. Through cooperative work of the multi-view diffusion model and the sparse view large-scale reconstruction model, rapid three-dimensional modeling from a single image is achieved, and the method has the advantages of being high in generation speed, high in model quality, rich in geometric details and the like.
Owner:YIKA (BEIJING) TECH CO LTD

Self-supervised single-view 3D reconstruction via semantic consistency

Apparatuses, systems, and techniques to identify a shape or camera pose of a three-dimensional object from a two-dimensional image of the object. In at least one embodiment, objects are identified in an image using one or more neural networks that have been trained on objects of a similar category and a three-dimensional mesh template.
Owner:NVIDIA CORP

Single-view large-scale outdoor scene three-dimensional reconstruction method based on three-dimensional Gaussian splashing

The invention discloses a single-view large-scale outdoor scene three-dimensional reconstruction method based on three-dimensional Gaussian splashing, and the method comprises the steps: collecting a pseudo aerial image, and constructing a panoramic multi-mode supervision end-to-end single-view three-dimensional reconstruction model; meanwhile, panoramic consistency supervision, semantic constraint depth regularization and a radial weighted luminosity loss and Gaussian cutting mechanism are introduced, so that the defect of insufficient geometric constraint of traditional single-view three-dimensional reconstruction is effectively overcome, and high-efficiency and high-fidelity three-dimensional modeling of a large-scale outdoor scene under single image input is realized; the method is suitable for various actual scenes such as smart city construction, automatic driving simulation, virtual reality / augmented reality, digital twinning and the like.
Owner:HANGZHOU MAQUAN INFORMATION TECH CO LTD

Dynamic scene cross-camera multi-target tracking method based on feature decoupling

The invention discloses a dynamic scene cross-camera multi-target tracking method based on feature decoupling. According to the method, multi-view synchronous input is allowed, and target features in each view are decoupled into local tracking query and global tracking query. For target tracking in each view angle, corresponding local tracking query is designed; local tracking queries may learn inter-frame features used to predict target locations and trajectories. Meanwhile, the same target in different view angles has a global tracking query which is used for cross-view-angle object feature interaction; the global tracking query utilizes an attention mechanism to promote information sharing between objects; according to the method, global distribution and discrimination loss are designed for global tracking query, so that different objects are distinguished while similar objects are associated between views. According to the method, different targets are distinguished in a single view angle, the consistency of the same target in different view angles is ensured, and the performance of more stable cross-camera multi-target tracking in a dynamic scene is realized.
Owner:ZHEJIANG UNIV

Single-view image three-dimensional modeling method and system based on comparative learning

The invention provides a single-view image three-dimensional modeling method and system based on comparative learning, and the method comprises the steps: starting from a single input image, generating a multi-view candidate image through a pre-training two-dimensional diffusion model in combination with score distillation sampling; performing positive and negative sample division on the candidate image pairs based on texture fidelity and structure coherence; constructing a perceptual contrast learning mechanism, calculating perceptual loss between the images, and optimizing texture consistency in a three-dimensional content generation process; proposing a number-aware triple loss function, and dynamically adjusting the contribution ratio of positive and negative samples in training; then, a super-resolution reconstruction module is embedded to perform detail enhancement on the image, and the quality of a contrast gradient signal is improved; and finally, performing joint training on the perception comparison loss and a neural radiation field or volume rendering module to generate a high-quality three-dimensional grid or volume rendering result. The problems that in an existing diffusion generation method, multi-view textures are inconsistent, structure reconstruction is not coherent, and the semantic distinguishing capacity is insufficient are effectively solved.
Owner:SUZHOU MODAL JUMP INTELLIGENT TECHNOLOGY CO LTD

Structural deformation measurement method based on geometric correction of multi-view panoramic camera

The invention discloses a structural deformation measurement method based on geometric correction of a multi-view panoramic camera. A method for measuring bridge displacement by using a single multi-view panorama camera is provided, multi-view segmentation is performed on a panorama image according to a panorama camera equivalent model and a cube projection method, and independent calibration and correction are performed to obtain a distorted image capable of covering a 360-degree full-field range; therefore, a single panoramic camera is equivalent to a vertical four-camera system without overlapped view fields, so that the reference target in any direction can be used for displacement correction. The invention further provides a camera disturbance elimination method based on a multi-view circle center array target and pose transformation matrix purification optimization, and a coordinate system transformation matrix after purification optimization is obtained through a high-precision multi-feature point tracking algorithm from camera single-view moving target pose measurement. And obtaining a multi-view geometric correction theoretical model of any view of the measurement target and the reference target in a 360-degree range by utilizing a multi-view equivalent camera fixed connection relationship.
Owner:SOUTHEAST UNIV

Single-view three-dimensional dressed human body reconstruction method based on pixel and voxel feature fusion

The invention discloses a single-view three-dimensional dressed human body reconstruction method based on pixel and voxel feature fusion. The method comprises the steps of data set preprocessing, data set dividing, three-dimensional dressed human body reconstruction network construction, three-dimensional dressed human body reconstruction network training, three-dimensional dressed human body reconstruction network testing and reconstruction result displaying. In the pixel feature extraction network, the coordinate attention and the stacking hourglass network are combined, so that the pixel feature extraction precision is improved; in the voxel feature extraction network, a three-dimensional channel and space attention are introduced into the voxelization human body parameterization model network, and a multi-scale residual feature extraction module is combined to extract voxel features. And splicing the extracted pixels and voxel feature vectors, inputting the spliced pixels and voxel feature vectors into a multi-layer perceptron, and reconstructing a three-dimensional dressed human body model through Marking Cubes, thereby solving the technical problem of insufficient precision of the existing single-view three-dimensional dressed human body reconstruction method, and being applicable to the technical fields of artificial intelligence and computer vision.
Owner:SHAANXI NORMAL UNIV

3D gaussian diffusion for single-view reconstruction

Methods, devices, and processor-readable media for method for performing a 3D reconstruction from a single view image, comprising progressively denoising a randomly initialized set of 3D-gaussian representations with continuous guidance from an input image.
Owner:HUAWEI TECH CO LTD

Rapid automatic focusing module and method suitable for fluorescence microscope

The invention discloses a rapid automatic focusing module and method suitable for a fluorescence microscope. The automatic focusing module comprises a laser, a diaphragm, a collimating lens group, a reflector, a dichroic mirror, an objective lens, a sample to be measured, a telescope lens group, a CCD2 detector, a control system and a displacement platform which are sequentially arranged along a focusing light path, and the sample to be measured, the objective lens, the dichroic mirror, a tube lens and the CCD2 detector which are sequentially arranged along an imaging light path; the collimated laser beam is emitted to the plane where the sample to be detected is located through the objective lens, and then the laser beam is reflected back to the telephoto lens group and converged to the CCD2 detector. According to the offset of the center of the reflected light beam, the relative distance between the objective lens and the to-be-tested sample is moved through a control system, and automatic focusing of a single view field of the to-be-tested sample is achieved; the to-be-measured sample is moved to the focal planes of different object view fields in the X direction and the Y direction and is quickly focused in real time, so that the large-size to-be-measured sample is always kept at the focal plane position, and the image quality and the working efficiency are improved.
Owner:NANJING WAVELENGTH OPTO ELECTRONICS SCI & TECH CO LTD

Multi-view conversion method based on multi-view consistency diffusion model

The invention relates to the technical field of image conversion, and discloses a multi-view conversion method based on a multi-view consistency diffusion model, and the multi-view consistency diffusion model is trained through a unified multi-view-to-multi-view conversion framework based on the diffusion model. The conversion framework has a new self-supervised training strategy and a consistency and antagonism supervision strategy, and the strategies utilize an existing 2D single-view image converter and a 3D Gaussian splash technology to generate pseudo-truth values as supervision signals, so that view consistency and high-frequency details are enhanced; the multi-view consistency diffusion model trained by the conversion framework further comprises a potential multi-view consistency module, and the potential 3DGS module is used as a 3D representation of a bottom layer to promote information exchange across multiple views, so that 3D priori is introduced into the diffusion model to ensure view consistency. Finally, a joint optimization strategy is provided, and better balance between consistency and authenticity can be realized.
Owner:UNIV OF SCI & TECH OF CHINA

Robot environment sensing method based on single-view three-dimensional scene generation

The invention relates to a robot environment perception method based on single-view three-dimensional scene generation, and the method comprises the steps: collecting a two-dimensional image containing target environment information through employing a monocular camera, generating multi-view information through combining depth estimation, normal prediction and a two-stage semantic guidance diffusion model, and constructing a high-quality three-dimensional scene. And reconstructing three-dimensional point cloud data through the neural radiation field, and performing texture rendering optimization on the point cloud data. A Point Net + + network is used for carrying out semantic analysis on point clouds, a multi-frame time sequence point cloud registration and Kalman filtering tracking method is introduced, modeling is carried out on a dynamic target, and a dynamic semantic map with a motion state is constructed. And finally, structured output environment information is used for robot navigation, path planning and task execution. The problems of high cost, high complexity and insufficient real-time performance and robustness of a three-dimensional scene generation technology in the field of robot environment perception are solved, and the method is high in structuring degree, standard and unified in output format and high in universality and engineering adaptation capacity.
Owner:DONGHUA UNIV

Three-dimensional model retrieval method based on multi-view aggregation

The invention provides a three-dimensional model retrieval method based on multi-view aggregation, and the method comprises the steps: 1, designing a three-dimensional model retrieval network based on views, and enabling the three-dimensional model retrieval network to extract the features of a three-dimensional model from the view representation of the three-dimensional model for three-dimensional model retrieval; the three-dimensional model retrieval network MVAT comprises a single-view feature extraction network and a multi-view feature aggregation network, the single-view feature extraction network is responsible for independently extracting local features of a three-dimensional model from each view, and the multi-view feature aggregation network is responsible for generating global feature representation through information fusion of multiple views; 2, weight freezing and contrast learning technologies are introduced, and a three-stage training method is adopted to train the three-dimensional model retrieval network; and 3, performing performance evaluation on the trained three-dimensional model retrieval network on a test set. The three-dimensional model retrieval method has the advantages of being high in feature expression ability, efficient in training method, high in precision and good in generalization.
Owner:HARBIN INSTITUTE OF TECHNOLOGY (SHENZHEN) (INSTITUTE OF SCIENCE AND TECHNOLOGY INNOVATION HARBIN INSTITUTE OF TECHNOLOGY SHENZHEN)

Cultural relic handicraft three-dimensional reconstruction method based on multiple views and deep learning

The invention discloses a cultural relic handicraft three-dimensional reconstruction method based on multiple views and deep learning, and relates to the technical field of cultural relic handicraft three-dimensional reconstruction methods, and the method comprises the steps: carrying out the image collection of cultural relic handicrafts through employing a multi-view shooting method, and obtaining a structured image data set; performing feature matching and camera parameter estimation on the obtained multi-view image set by adopting a sparse reconstruction method to obtain camera internal parameters, external parameters and sparse point cloud data; performing fine processing on the obtained sparse point cloud data by adopting a dense reconstruction method to obtain a 3D dense point cloud model with a texture map; analyzing the obtained 3D dense point cloud model by adopting a depth map generation method to obtain a depth map and a rendering image corresponding to each image; image acquisition is performed on the cultural relic handicraft by adopting a multi-view shooting method, so that information such as geometric shapes and texture details of the cultural relic handicraft is comprehensively captured, and the problem of information loss caused by a single view angle is effectively solved.
Owner:QINGHAI NORMAL UNIV

Single-view human body three-dimensional reconstruction method based on Gaussian surface elements

The invention discloses a single-view human body three-dimensional reconstruction method based on Gaussian surface elements, and the method comprises the steps: collecting image data from a plurality of human body data sets, carrying out the preprocessing through employing a targeted preprocessing strategy according to the data characteristics corresponding to each image data, and obtaining a processed human body data set; constructing a human body 3D Gaussian attribute prediction model, wherein the human body 3D Gaussian attribute prediction model comprises a hierarchical feature extraction module, a feature fusion network and a 3D Gaussian function decoder; training the human body 3D Gaussian attribute prediction model based on the processed human body data set to obtain a trained human body 3D Gaussian attribute prediction model; and performing three-dimensional reconstruction on an input human body picture based on the trained human body 3D Gaussian attribute prediction model, and outputting a reconstructed human body three-dimensional image. According to the method, the manufacturing cost of the virtual digital human is greatly reduced, and the efficiency and accuracy of constructing the dynamic human body model are improved.
Owner:SHENZHEN UNIV

Vehicle positioning method based on multi-view and multi-scale feature fusion

The invention belongs to the technical field of vehicle positioning, and particularly discloses a vehicle positioning method based on multi-view and multi-scale feature fusion. According to the method, a multi-view and multi-scale pyramid fusion neural network is provided, at the input stage of the network, a multi-channel distance image RIV of laser radar point cloud projection and BEV representation generated through normal distribution transformation are utilized, mutual complementation can be achieved on the global scale and the local scale, the overall understanding ability of a scene is improved, and the overall performance of the scene is improved. The limitation of a single view angle is effectively relieved through multi-view-angle input, and the robustness and anti-interference performance of the model are improved. On the network architecture level, a self-attention mechanism using Mama is provided. According to the method, not only can the complementary correlation among multiple views be fully utilized, but also the context information can be coded and integrated across multi-scale features, the finally generated global descriptor features are rich, and the accuracy of vehicle positioning is improved.
Owner:SHANDONG UNIV OF SCI & TECH

Counter-defense method based on multi-view image super-resolution

The invention discloses a confrontation and defense method based on multi-view image super-resolution, and the method comprises the steps: obtaining an attacked image, and enabling the attacked image and a clean image to form a clean image and attacked image pair; inputting the attacked image and the auxiliary view angle image into a network for coding to obtain implicit representation; obtaining a homography matrix from an auxiliary view angle image to an attacked image, and performing spatial transformation on the auxiliary view angle image on a feature level to realize spatial alignment; implicit representation is fused in a feature space through a decoding function, a clean pixel value is obtained by polling each pixel position, and defensive output after the network purifies the attacked image is obtained. The method does not depend on the limitation of a single view angle image, reduces the dependence on the consistency of observation angles, effectively achieves the purification of adversarial disturbance, improves the visual quality of the image, enhances the defense capability of the network for potential adversarial attacks, and can be flexibly applied to an image processing robustness task in a complex real environment.
Owner:TIANJIN UNIV

Multi-view structure learning method based on multi-expert cooperation

The invention provides a multi-view structure learning method based on multi-expert cooperation, and relates to the technical field of graph neural networks and multi-view learning, and the method comprises the steps: reconstructing initial multi-view data, and obtaining multi-view structure data; respectively training the single-view expert model and the shared expert model by utilizing a first loss function and a second loss function based on the multi-view structure data to obtain a trained single-view expert model and a trained shared expert model; constructing a collaborative decision model by using the trained single-view expert model and the trained shared expert model; training the collaborative decision model by using a third loss function to obtain a trained collaborative decision model; and analyzing the graph data by using the trained collaborative decision model to obtain a joint decision result, and completing learning of the multi-view structure. According to the method, the problems of large structural noise interference, isolated expert model information and insufficient node classification accuracy when an existing graph neural network processes multi-relation graph data are solved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Crop plant three-dimensional dynamic model construction method and device based on digital twinning

The invention discloses a crop plant three-dimensional dynamic model construction method and device based on digital twinning, and the method comprises the steps: constructing a plant growth model database, and storing the size and relative position relation of each organ of a plant in different growth environments and different growth stages; obtaining the image of each organ of the plant, and generating a depth image of each organ of the plant through a single-view depth estimator; converting the depth image of each organ of the plant into a three-dimensional point cloud data set, extracting three-dimensional surface information of the organ from the three-dimensional point cloud data set, and then converting the three-dimensional surface information into unit spherical projection of the organ; using a generator to obtain a unit fine three-dimensional model of each organ according to the unit spherical projection; setting different growth environments and growth stages, extracting the size and relative position relationship of each organ of the plant from the plant growth model database, and performing scaling and splicing by taking the unit fine three-dimensional model of each organ as a reference to form an overall three-dimensional model of the plant. According to the method, the time consumed for building the whole plant model is short, and the obtained plant has good simulation.
Owner:HUNAN CREATOR INFORMATION TECH CO LTD

Single-view unknown object 6D pose estimation method based on segmentation and new view angle synthesis

The invention discloses a single-view unknown object 6D pose estimation method based on segmentation and new view angle synthesis, and relates to the field of object pose estimation. The method comprises the following steps: designing a segmentation module and a new view angle synthesis module of a pose estimation network, and obtaining a segmentation result of a target object by using the segmentation module; generating a target object multi-view-angle embedding template by using a new view angle synthesis module; matching the embedding representation zquery'of the segmentation result with a multi-view embedding template of the target object to find out the embedding representation most similar to zquery ', and determining a preliminary rotation attitude estimation result R of the target object; and estimating the relative translation vector t of the target object, and combining the R to obtain the relative pose (R, t) of the target object. According to the method, robust 6D pose estimation of the unknown object in a complex visual scene can be realized only by taking a single reference image as input, and the pose estimation precision and robustness of the unknown object in the complex scene are improved while the real-time requirement is met.
Owner:NORTHEASTERN UNIV CHINA

Single view reconstruction and rendering method

The invention discloses a single view reconstruction and rendering method, which belongs to the field of view reconstruction and rendering, and comprises the following steps of: firstly, generating multi-view feature representation with strong geometric consistency from a single input image by introducing an image diffusion module of a cross attention mechanism; a point cloud reconstruction module with self-attention and cross-attention is utilized, and multi-view information is fused to reconstruct an accurate three-dimensional point cloud; and finally, constructing differentiable three-dimensional Gaussian representation based on the point cloud, rendering the differentiable three-dimensional Gaussian representation, and outputting a new view angle image, a normal map and a depth map. According to the system, a staged training strategy is adopted, and a composite loss function including multi-scale bidirectional consistency smooth loss and feature consistency loss is innovatively used for optimization. According to the method, the problems of low geometric accuracy, multi-view inconsistency, detail missing and the like in single-view reconstruction are effectively solved, and the method can be widely applied to the fields of virtual reality, digital twinning, cultural heritage digitization and the like.
Owner:北京渲光科技有限公司

Layer focusing illustration single-view 3D reconstruction method and system

The invention provides a layer focusing illustration single-view 3D reconstruction method, which comprises the following steps of: performing region segmentation on an original illustration single view in a data set through SAM to obtain a region text pair fine-tuning CLIP model comprising the original illustration single view and a layer channel of the original illustration single view; extracting an unnatural contour line, semantic information and layer information from each original illustration single view in the data set T; selecting a basic element model to initialize a 3D point cloud according to a role posture in a single view, segmenting the human body point cloud into different parts through point cloud segmentation, carrying out noise growth and color perturbation on corresponding point cloud regions according to position information of an unnatural contour line, respectively mapping the point cloud regions into a submerged space, and obtaining a final potential representation by adopting LDM; and finally, the final potential representation passes through a decoder to obtain a final 3D representation, and 3D assets are reconstructed. According to the invention, the corresponding 3D assets can be generated according to a given illustration single view.
Owner:XIANGTAN UNIV

Deep sea polymetallic nodule single view three-dimensional reconstruction method and system

The invention belongs to the technical field of image processing and deep-sea mining, and discloses a deep-sea polymetallic nodule single-view three-dimensional reconstruction method and system. Aiming at the problem that the real three-dimensional data of the underwater polymetallic nodule is difficult to acquire, a multi-view generation model is adopted as 2D prior, 3D Gaussian is optimized through fractional distillation sampling loss, any three-dimensional data training is not used, the problem of three-dimensional data acquisition is solved, and three-dimensional reconstruction of the polymetallic nodule is realized. In addition, the invention provides a method only through image fine tuning, function decoupling of a noise prediction module (Unet) and a condition control network (multi-view model) is realized based on DreamBooth, and multi-view generation stability can still be maintained through pre-training view angle parameters under empty condition input; and the problem of parameter acquisition of the underwater camera is avoided through a conditional blanking mechanism, so that the model can be finely adjusted only by depending on a single-view image, and the data acquisition cost is reduced.
Owner:SHANDONG UNIV OF SCI & TECH

Multi-view image generation method and multi-view image generation model training method

The invention provides a multi-view image generation method and a multi-view image generation model training method. The multi-view-angle image generation method comprises the steps that a target single-view-angle image of a target object is obtained, the target single-view-angle image is input into a pre-trained multi-view-angle image generation model, m-view-angle images of the target object are output, the image of one view angle comprises n-domain images, m and n are integers larger than 1, and m is an integer larger than 1. The images of different domains of the target object under the same view angle have consistency in physical attributes. A single view angle can be expanded to multiple view angles, and richer visual information is provided; moreover, the physical attributes of the domain images in the multi-view image are consistent, so that vivid illumination and material performance can be ensured, and the details of the generated image are richer and more accurate.
Owner:ALIPAY (HANGZHOU) INFORMATION TECH CO LTD

Lane detection and distance estimation using single-view geometry

ActiveUS12175695B2Image enhancementMathematical modelsInverse perspective mappingComputer graphics (images)
Disclosed are methods, devices, and computer-readable media for detecting lanes and objects in image frames of a monocular camera. In one embodiment, a method is disclosed comprising receiving a sample set of image frames; detecting a plurality of markers in the sample set of image frames using a convolutional neural network (CNN); fitting lines based on the plurality of markers; detecting a plurality of vanishing points based on the lines; identifying a best fitting horizon for the sample set of image frames via a RANSAC algorithm; computing an inverse perspective mapping (IPM) based on the best fitting horizon; and computing a lane width estimate based on the sample set of image frames using the IPM in a rectified view and the parallel line fitting.
Owner:MOTIVE TECHNOLOGIES INC