28 results about "Appearance based" patented technology

One example provides a computing device comprising a display device, a logic subsystem, and a memory subsystem. The memory subsystem comprises instructions that are executable by the logic subsystem to display a graphical airspace representing an airspace on the display device, and obtain one or more notice to air missions (NOTAMs) related to the airspace. The instructions are further executable to, for each NOTAM of the one or more NOTAMs, display a graphic overlay on the graphical airspace. The graphic overlay has an appearance based at least upon a determined status of the NOTAM.

Methods, systems, and apparatuses, among other things, may label and classify objects via appearance-based clustering for computer vision automatic target recognition (ATR) systems, including automated anomaly detection for objects appearing in an area of interest (AOI).

The present disclosure generally relates to methods and user interfaces for establishing communications. A computer system displays a user interface object corresponding to a communication with different appearances based on a respective state of a process for sending the communication.

Disclosed is a method, system and device for estimating the three-dimensional gaze direction of a human eye based on appearance. The method comprises: acquiring an image or video containing a face; performing face detection in the image or video frame, introducing a face key point detection algorithm to detect and locate the face key points; cutting out the face image and left and right eye images according to the face detection result and the positions of the key points; and estimating the three-dimensional gaze of the human eye in the image using a gaze estimation model, wherein the training process of the gaze estimation model comprises: using a Base-CNNs network to extract features of the face image to obtain a full-face image feature vector, using two asymmetric branch structures of an EE-Net network to extract features of the left and right eye images to obtain double-eye image feature vectors, estimating a head posture vector in combination with the face key point detection algorithm, and finally using a full connection layer to regress the full-face image feature vector, the double-eye image feature vector and the head posture vector for the final gaze estimation task.

The appearance of an object is evaluated using at least one of a single type of data and multiple types of data. [Solution] The information processing system inputs appearance data representing the appearance of an object into a first base model, which is a trained model capable of processing at least one of a single type of data and multiple types of data, thereby obtaining appearance base data output from the first base model. Based on the appearance base data, the information processing system obtains related data associated with the appearance base data from multiple base data stored in a database. The information processing system inputs the appearance base data and related data into a second base model, which is a trained model, thereby obtaining evaluation data representing the evaluation of the object's appearance, and outputs the evaluation data.

The invention provides an intelligent quality inspection system and method for a precast beam field, electronic equipment and a storage medium. The intelligent quality inspection system comprises a conveying and positioning subsystem, a detection subsystem and a control subsystem, wherein the conveying and positioning subsystem is used for conveying a to-be-detected prefabricated part to a detection station; the central processing unit is used for sending a detection task instruction, performing multi-dimensional comprehensive analysis processing based on an appearance quality identification result, a geometric dimension identification result and an internal defect identification result, and determining a comprehensive quality evaluation result; the appearance quality detection subsystem is used for carrying out appearance defect detection processing on the surface image information based on an artificial intelligence visual identification algorithm and determining an appearance quality identification result; the geometric dimension detection subsystem is used for performing dimension analysis on the actual geometric parameters to determine a geometric dimension identification result; and the internal defect detection subsystem is used for processing the internal structure information based on an internal defect analysis algorithm and determining an internal defect identification result. The quality inspection efficiency is greatly improved, and the efficiency bottleneck problem of traditional manual detection is solved.

The application relates to the technical field of robots, in particular to a lightweight topological map construction method for scene recognition, which comprises the following steps: collecting data and processing the data to obtain accumulated point cloud; extracting scene descriptors; calculating the similarity between the descriptors to generate topological nodes; and calculating an optimal topological map. The application describes a scene according to multi-frame point cloud data accumulated according to a robot running track, extracts a descriptor of the point cloud, and judges whether to generate a topological node according to the similarity of the descriptor; the application can guarantee effective construction of a lightweight topological map, is particularly suitable for the demand of large-range scene topological map construction, and compared with an appearance-based topological map construction method, the number of topological map nodes constructed by the application is greatly reduced, but the reliable scene recognition performance can still be guaranteed.

The disclosure provides an image repairing method and system combining edge information and appearance flow operation, the method comprising: preprocessing an obtained image to be repaired to obtain a smooth image; performing edge detection and prediction to obtain a target edge map; performing smooth structure reconstruction according to the smooth image and the target edge map to obtain reconstructed smooth structure; extracting the correlation between each region of the image to be repaired according to the reconstructed smooth structure, sampling features from the relevant regions based on the appearance flow operation to generate lively texture, and filling the features containing lively texture information to the damaged region of the image to be repaired to obtain a repaired complete image. The disclosure realizes the repair of the damaged image through the repair process from the edge to the smooth structure and then to the whole, combines the edge information and the appearance flow operation to perform three-stage image repair, can generate reasonable structure and real lively texture, and thus achieves good repair effect.

The invention relates to the technical field of image acquisition, in particular to a railway vehicle appearance image acquisition system and method and device.The system comprises a movable support, an image acquisition module and a control module, and the image acquisition module is arranged on the movable support and used for acquiring an appearance image of a target component of a railway vehicle; the movable bracket is in communication connection with the control module and is used for driving the image acquisition module to move to a position corresponding to a position moving instruction based on the position moving instruction issued by the control module; the control module is used for receiving an appearance image acquisition instruction, issued by a user, of the target component of the railway vehicle, and forming a position moving instruction based on the appearance image acquisition instruction, so that the image acquisition module can acquire an appearance image of the target component of the railway vehicle after the movable bracket moves to a position corresponding to the position moving instruction; the appearance image of the railway vehicle can be dynamically collected according to user requirements, and personalized and refined detection requirements are met.

The present application relates to the technical field of image processing, and discloses a gait recognition method, which captures appearance contour information from different angle positions by polar coordinate sampling at R different sampling angles, explicitly models contour structure in a polar coordinate system, makes up for the deficiency of traditional appearance-based methods in capturing structural features, and thus improves the robustness of gait recognition to appearance changes. Furthermore, the present application also performs multi-scale fusion on structural features of different sampling angles to suppress redundant information and enhance the stability and distinguishability of structural representation. In addition, the present application combines gait features with global information and local information on the final structural features obtained by polar coordinate sampling, maintains the robustness to appearance disturbance, and improves the sensitivity of the model to shape changes. The present application effectively alleviates the influence of view angle change, clothing change and carrying articles, and greatly improves the robustness of appearance contour modeling.

The invention relates to the technical field of PCB (Printed Circuit Board) detection, in particular to a method and a system for detecting the edge deviation of the appearance of a PCB based on a welding component. The polarization characteristic pattern is generated by acquiring the multi-spectral original image and the polarization state parameter of the to-be-detected area of the PCB, the reflection exposure abnormal area and the reflection exposure normal area can be effectively positioned, the fusion weight is automatically adjusted according to the spectral response characteristic of the signal area through the self-adaptive spectrum fusion enhancement technology, and the detection accuracy is improved. The problem that in a traditional method, image enhancement is not fine enough is solved, the adaptive fusion enhancement technology enables the multispectral image to better conform to the actual situation during enhancement, more detail information is reserved, key features can be accurately extracted by generating a semantic segmentation map and combining three-dimensional standard coordinate information, and the image enhancement accuracy is improved. The detection precision and reliability of the welding component and the appearance edge of the PCB are effectively improved, the deviation can be accurately positioned and quantified, and the quality control level in the production process of the PCB is improved.

We provide the system. [Solution] A means of acquiring images of the user using a video acquisition device, A means for identifying a face and extracting its appearance based on acquired image data, A means of presenting a style suitable for the user using an AI model via an information processing device, based on extracted appearance data, A means of virtually applying the presented style to an image in real time and providing it to the user using a display device, A means of presenting product information related to the selected style and providing a purchase interface, A means of providing an interface that enables users to make purchase decisions based on the presented virtual style, A system that includes this.

Proposed are a method and apparatus including real-time eye tracking and calibration functions in a kiosk including a display. The method may include collecting first data related to a user's appearance based on image data on the user acquired from a camera of the kiosk, and generating first screen coordinates of the display corresponding to the user's eyes based on the first data and performing eye tracking on the user. The method may also include collecting second data including second screen coordinates for a corresponding touch point detected when the user touches and selects a menu displayed on the display and training a calibration machine learning model using the collected second data, and performing inference on the calibration model using the first screen coordinates according to the eye tracking performed at a time other than the touch selection time point to perform calibration on the eye tracking.

An object of a system according to an embodiment is to predict the risk of appearance of bears and promptly notify residents of the risk.SOLUTION: A system includes an action analysis part, a prediction part, and a notification part. The behavior analysis unit analyzes a behavior pattern of the bear. The prediction unit predicts a risk of appearance of a bear on the basis of the behavior pattern of the bear analyzed by the behavior analysis unit. The notification unit notifies the resident of the risk of appearance of the dark circle predicted by the prediction unit.SELECTED DRAWING: Figure 1

The application provides a head pose estimation method and system, the method comprising: acquiring a two-dimensional face image to be recognized; performing face calibration on the two-dimensional face image to obtain a calibrated two-dimensional face image; acquiring face feature points on the calibrated two-dimensional face image; inputting the face feature points into a regression network to obtain a 6D rotation matrix, the regression network being a regression model based on 6D rotation representation; converting the 6D rotation matrix into a 3*3 rotation matrix according to orthogonal mapping; and converting the 3*3 rotation matrix into Euler angles, the Euler angles being used to represent the rotation angle of the head pose, so as to solve the problem that the recognition accuracy and robustness of the model-based pose estimation method and the appearance-based pose estimation method cannot be accurately represented at the same time.

The invention discloses an intervention device full-process tracing and abnormity identification method based on RFID and image identification, and belongs to the technical field of medical device informatization management. The method comprises the following steps: allocating an RFID tag to an interventional device, and deploying a reader-writer and an image acquisition device in each link; generating appearance fingerprint information based on the appearance image; in warehousing, preoperative, intraoperative and postoperative links, by comparing appearance fingerprints, checking validity periods and analyzing positioning tracks and operator identities, verification results of all the links are obtained, and cross-link tracing data is formed; performing hierarchical processing on the data to obtain an identification result of the basic anomaly and the associated anomaly; and associating the traceability data and the abnormal result with the digital twin identifier, and storing the associated data and the associated result into a block chain. The method solves the technical problems that in the prior art, intervention equipment tracing information is isolated and tampered easily, equipment replacement cannot be recognized in a cross-link mode, and the use process is abnormal.

This application provides a method, apparatus, computer device, computer-readable storage medium, and computer program product for controlling the display of a virtual avatar, belonging to the field of computer technology. The method includes: receiving a gift-giving event targeting a target virtual avatar; generating a corresponding data object based on the gift-giving event, wherein the data object includes capability attribute information, which characterizes the effect of the gift-giving event on the appearance of the target virtual avatar; and outputting a control result for controlling the display of the target virtual avatar's appearance based on multiple data objects. The technical solution of this application can achieve the superposition, synergy, or enhanced expression of the effects of multiple gift-giving events on the appearance of the target virtual avatar, improving the gift-giving display effect.

The invention relates to a small sample precast concrete defect image generation method and system, and the method comprises the steps: obtaining defect images and defect-free images of different textures, and processing the defect images, thereby obtaining a small sample defect image training set and a normal image data set; based on the small sample defect image training set, separating appearance features and spatial features of defect images and defects according to a stable diffusion model in combination with a control network model to obtain appearance feature embedding vectors; smoothly migrating the general knowledge to the precast concrete defect field according to a low-rank adaptive parameter fine tuning algorithm, and generating a precast concrete defect mask; and controlling the generated defect appearance content according to the feature learning model based on the appearance feature embedding vector, taking the normal image data and the defect mask as input data in combination with the control network model, controlling the position of the generated defect appearance, and generating a large number of high-quality defect images. According to the method and the device, diversified high-quality precast concrete defect images can be generated by using a very small number of samples.

The invention discloses a food package detection and classification method and system based on machine vision, and relates to the technical field of food package detection.The method comprises the steps that sample images of various types of food packages are obtained, feature extraction is conducted on each sample image, and a feature cloud cluster of the food packages is generated and stored in an incremental database; the method comprises the following steps: acquiring a detection image of a to-be-detected food package based on an industrial camera, and performing feature extraction on the to-be-detected image by using a dual-channel detection mechanism to obtain an appearance feature description matrix and a spatial structure topological graph; based on the appearance feature description matrix and the spatial structure topological graph, feature data in the incremental database are matched, the food package category in the to-be-detected image is determined, a sorting instruction is generated, the sorting instruction is transmitted to a PLC control module, and the PLC module executes sorting operation based on the sorting instruction; through an innovative two-channel feature extraction mechanism and a feature cloud cluster technology, the practicability and reliability of the food package detection method in a real industrial environment are greatly improved.

This disclosure generally relates to presenting avatars in three-dimensional environments. In some embodiments, a computer system receives data indicative of a pose of at least a first portion of a user, and causes presentation of an avatar comprising respective avatar features corresponding to the first portion of the user and presented with variable display properties indicative of a certainty of the pose of the first portion of the user. In some embodiments, a computer system receives data indicative that a current activity of one or more users is a first type of activity, and in response, updates a representation of a user with a first appearance based on a first appearance template. The system receives second data indicative of a current activity of the one or more users, and in response, updates the appearance of the representation of the first user based on the current activity of the one or more users using the first or second appearance template.

This invention discloses a visual localization and recognition method and apparatus based on a joint appearance and semantic descriptor, relating to the field of image recognition technology, with the main objective of solving the problem of low accuracy in localization and recognition. The method mainly includes: extracting pixel-level semantic segmentation maps of the image to be recognized; segmenting semantic objects based on the pixel-level semantic segmentation maps to obtain multiple semantic objects to be recognized; performing semantic structure modeling based on the semantic contour information of all semantic objects to be recognized to obtain a global semantic descriptor, and selecting multiple candidate images that match the global semantic descriptor; extracting the joint semantic appearance descriptor corresponding to the image to be recognized and different candidate images respectively, and determining the target image from the candidate images based on the similarity to the joint semantic appearance descriptor of the image to be recognized; and calculating the pose information of the object to be located based on the feature point matching relationship between the target image and the image to be recognized. This method is mainly used for visual localization.

The application provides a three-dimensional Gaussian splash appearance robustness enhancement method and system based on generative latent optimization, assigns an independent image appearance latent vector for each image in a training image set for encoding global appearance attributes of the image, assigns an independent Gaussian appearance latent vector for each Gaussian particle in a three-dimensional scene for encoding adjustable appearance features of the particle, generates a color adjustment amount and an opacity adjustment amount for the Gaussian particle based on the image appearance latent vector corresponding to a target image and the Gaussian appearance latent vector of the target Gaussian particle through a transformation network, modifies rendering attributes of the target Gaussian particle according to the color adjustment amount and the opacity adjustment amount, and inputs the adjusted Gaussian particle into a rasterization renderer to generate a target view image. The application enables precise appearance adaptation of microscopic regions such as local shadow boundaries and highlight reflection regions, and eliminates artifacts caused by global parameter overfitting from a mechanism.

A road sign detection method comprises obtaining an unlabeled image of a road segment and executing a neural network trained to identify road signs within the unlabeled image. The neural network comprises a first model configured to localize an object in the unlabeled image. The neural network may further comprise a second model configured to generate classification data associated with the cropped image, based on content of the cropped image, where the classification data corresponds to one or more appearance-based attributes of the cropped image and also to determine a category of the cropped image, based on the classification data of the cropped image. The neural network may comprise a third model configured to generate road sign label data based on the determined category of the cropped image, where the road sign label data represents the identified road sign. The method further comprises outputting the road sign label data.

The invention provides a three-dimensional Gaussian splash appearance robustness enhancement method and system based on generative potential optimization, and the method comprises the steps: distributing an independent image appearance latent vector for each image in a training image set, and enabling the image appearance latent vector to be used for coding the global appearance attribute of the image; distributing an independent Gaussian appearance latent vector for each Gaussian particle in the three-dimensional scene, wherein the Gaussian appearance latent vector is used for encoding adjustable appearance features of the particles; based on the image appearance latent vector corresponding to the target image and the Gaussian appearance latent vector of the target Gaussian particle, generating a color adjustment amount and an opacity adjustment amount for the Gaussian particle through a transformation network; and modifying the rendering attribute of the target Gaussian particle according to the color adjustment amount and the opacity adjustment amount, and inputting the adjusted Gaussian particle into a rasterization renderer to generate a target visual angle image. According to the method, microscopic areas such as local shadow boundaries and highlight reflection areas are subjected to accurate appearance adaptation, and the problem of artifacts caused by global parameter overfitting is eliminated from the aspect of mechanism.

The application relates to a small sample prefabricated concrete defect image generation method and system, which comprises the following steps: acquiring a defect image and a non-defect image with different textures, processing the defect image, obtaining a small sample defect image training set and a normal image data set; based on the small sample defect image training set, appearance features and spatial features of the defect image are separated according to a stable diffusion model combined with a control net model, and an appearance feature embedding vector is obtained; general knowledge is smoothly migrated to the prefabricated concrete defect field according to a low-rank adaptive parameter fine-tuning algorithm, and a prefabricated concrete defect mask is generated; based on the appearance feature embedding vector, the generated defect appearance content is controlled according to a feature learning model, the normal image data and the defect mask are taken as input data in combination with the control net model, the position of the generated defect appearance is controlled, and a large number of high-quality defect images are generated. The application can generate high-quality prefabricated concrete defect images with diversity by using a small amount of samples.

The invention discloses a video instance segmentation method and system based on a track guide memory network, and relates to the technical field of computer vision. According to the method, a trajectory-appearance joint modeling module is designed, a learnable threshold gating layer is introduced into a time sequence state modeling branch to filter a noise trajectory, and trajectory features and appearance features are fused through a mask self-attention mechanism, so that robust instance query is generated. On the basis, a time sequence perception memory updating and management module is designed, and historical features are denoised through a channel-space collaborative filtering mechanism. And dynamically updating a memory pool with a fixed size based on the appearance similarity, maintaining the feature quality of the memory pool, and introducing a time sequence position code to endow historical features with a time sequence perception capability. According to the method, high-precision segmentation and cross-frame identity association of the video instance in a complex scene are realized, and the problem of instance identity association error caused by lack of motion trail modeling and memory noise accumulation is effectively relieved.

An object of a system according to an embodiment is to make it possible to intuitively grasp a cause of a body shape change or a weight increase of a user.SOLUTION: A system includes a registration part, an input part, a simulation part, a calorie input part, a prediction part, a linkage part, and an analysis part. The registration unit registers a height, a weight, and a photograph of a user. The input unit measures and inputs a daily weight. The simulation unit simulates a change in appearance based on the data collected by the registration unit and the input unit. The calorie input section is used to input daily intake calories or nutrients. The prediction unit predicts a body shape after one year based on the data collected by the calorie input unit. The interlocking unit interlocks with the genetic test result. The analysis unit analyzes the cause of the weight gain.SELECTED DRAWING: Figure 1

The invention relates to an ancient building protection and text and travel resource development system based on digital twinning, and the system comprises a data collection module which collects the appearance image and spatial data of an ancient building; the digital twin model construction module is used for constructing an initial historic building three-dimensional model based on the appearance image and the spatial data, and generating a high-precision digital twin historic building three-dimensional model through a real-time rendering engine and parameter adjustment; the application service module is used for loading the three-dimensional model and providing interactive browsing of roaming at the perspective of a first person and aerial view at the perspective of an unmanned aerial vehicle; real-time measurement of component length or space area is carried out on the digital twin ancient building three-dimensional model; and providing a cultural and creative product database, a tour route database and a research course database of the historic building. Therefore, the problems of lack of integrated management of geometric morphology and multi-modal cultural assets of the ancient building, too low dimension of digital archive information, limitation of ancient building display and the like are solved, and intellectualization and scientization of protection, management and inheritance of cultural relics of the ancient building are realized.