4940results about "Three-dimensional object recognition" patented technology

A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of a first object detected in a first group of the images. The first object has a first signature representing features of the first object. The match classifier matches object instances by analyzing data derived from the first signature of the first object and a second signature of a second object detected in a second image. The second signature represents features of the second object derived from the second image. The match classifier determine whether the second signature matches the first signature. A training process automatically configures the match classifier using a set of possible object features.

In some arrangements, product packaging is digitally watermarked over most of its extent to facilitate high-throughput item identification at retail checkouts. Imagery captured by conventional or plenoptic cameras can be processed (e.g., by GPUs) to derive several different perspective-transformed views—further minimizing the need to manually reposition items for identification. Crinkles and other deformations in product packaging can be optically sensed, allowing such surfaces to be virtually flattened to aid identification. Piles of items can be 3D-modelled and virtually segmented into geometric primitives to aid identification, and to discover locations of obscured items. Other data (e.g., including data from sensors in aisles, shelves and carts, and gaze tracking for clues about visual saliency) can be used in assessing identification hypotheses about an item. A great variety of other features and arrangements are also detailed.

The technology of the 4D-GIS system deploys a GIS-based algorithm used to determine the location of a moving target through registering the terrain image obtained from a Moving Target Indication (MTI) sensor or small Unmanned Aerial Vehicle (UAV) camera with the digital map from GIS. For motion prediction the target state is estimated using an Extended Kalman Filter (EKF). In order to enhance the prediction of the moving target's trajectory a fuzzy logic reasoning algorithm is used to estimate the destination of a moving target through synthesizing data from GIS, target statistics, tactics and other past experience derived information, such as, likely moving direction of targets in correlation with the nature of the terrain and surmised mission.

A method and apparatus for estimating poses of a subject by grouping data points generated by a depth image into groups representing labeled parts of the subject, and then fitting a model representing the subject to the data points using the grouping of the data points. The grouping of the data points is performed by grouping the data points to segments based on proximity of the data points, and then using constraint conditions to assign the segments to the labeled parts. The model is fitted to the data points by using the grouping of the data points to the labeled parts.

Methods and arrangements involving portable user devices such smartphones and wearable electronic devices are disclosed, as well as other devices and sensors distributed within an ambient environment. Some arrangements enable a user to perform an object recognition process in a computationally- and time-efficient manner. Other arrangements enable users and other entities to, either individually or cooperatively, register or enroll physical objects into one or more object registries on which an object recognition process can be performed. Still other arrangements enable users and other entities to, either individually or cooperatively, associate registered or enrolled objects with one or more items of metadata. A great variety of other features and arrangements are also detailed.

Image analysis techniques, including object recognition analysis, are applied to images obtained by one or more image capture devices deployed within inventory environments. The object recognition analysis provides object recognition data (that may include one or more recognized product instances) based on stored product (training) images. In turn, a variety of functionalities may be enabled based on the object recognition data. For example, a planogram may be extracted and compared to a target planogram, or at least one product display parameter for a product can be determined and used to assess presence of the product within the inventory environment, or to determine compliance of display of the product with a promotional objective. In yet another embodiment, comparisons may be made within a single image or between multiple images over time to detect potential conditions requiring response. In this manner, efficiency and effectiveness of many previously manually-implemented tasks may be improved.

A computer-implemented method is provided to automatically apply predefined privileges for identified and tracked users in a space having one or more media sources. The method includes an operation to define and save to memory, a user profile. The user profile may include data that identifies and tracks a user with a depth-sensing camera. In another operation privileges that define levels of access to particular media for the user profile are defined and saved. The method also includes an operation to capture image and depth data from the depth-sensing camera of a scene within the space. In yet another operation, the user is tracked and identified within the scene from the image and depth data. In still another operation the defined privileges are automatically applied to one or more media sources, so that the user is granted access to selected content from the one or more media sources when identified and tracked within the scene.

A method and apparatus for performing two-dimensional video alignment onto three-dimensional point clouds. The system recovers camera pose from camera video, determines a depth map, converts the depth map to a Euclidean video point cloud, and registers two-dimensional video to the three-dimensional point clouds.

An apparatus, method for detecting critical areas and a pedestrian detection apparatus using the same are provided. An application of the pedestrian detection system is provided to help limit critical urban environment to particular areas. Contrary to traditional pedestrian detection systems that localize every pedestrians appearing in front of the subject vehicle, the apparatus first finds critical areas from urban environment and performs a focused search of pedestrians. The environment is reconstructed using a standard laser scanner but the subsequent checking for the presence of pedestrians is performed by incorporating a vision system. The apparatus identifies pedestrians within substantially limited image areas and results in boosts of timing performance, since no evaluation of critical degrees is necessary until an actual pedestrian is informed to the driver or onboard computer.

A method and system analyzes data acquired by image systems to more rapidly identify objects of interest in the data. In one embodiment, z-depth data are segmented such that neighboring image pixels having similar z-depths are given a common label. Blobs, or groups of pixels with a same label, may be defined to correspond to different objects. Blobs preferably are modeled as primitives to more rapidly identify objects in the acquired image. In some embodiments, a modified connected component analysis is carried out where image pixels are pre-grouped into regions of different depth values preferably using a depth value histogram. The histogram is divided into regions and image cluster centers are determined. A depth group value image containing blobs is obtained, with each pixel being assigned to one of the depth groups.

A processor-based system operating according to digitally-embedded programming instructions performs a method including identifying a group of pixels corresponding to a face region within digital image data acquired by an image acquisition device. A set of face analysis parameter values is extacted from said face region, including a faceprint associated with the face region. First and second reference faceprints are determined for a person using reference images captured respectively in predetermined face-portrait conditions and using ambient conditions. The faceprints are analyzed to determine a baseline faceprint and a range of variability from the baseline associated with the person. Results of the analyzing are stored and used in subsequent recognition of the person in a subsequent image acquired under ambient conditions.

Aspects of the present invention relate to context based power management of a head-worn computer.

A method for three-dimensional shape and size measurement of a three-dimensional body surface comprising providing a three-dimensional scanner, providing a processor, providing a three-dimensional Computer Aided Design system, scanning in three dimensions with the three-dimensional scanner at least a portion of the three-dimensional body surface, creating a data file representative of the three-dimensional body surface, processing the data file with the processor, importing the data file into the three-dimensional Computer Aided Design System, employing the three-dimensional Computer Aided Design System relative to the data file to define and record three-dimensional measuring data relative to at least a portion of the three-dimensional body surface, A and employing the three-dimensional Computer Aided Design System to exploit the three-dimensional measuring data.

A system and method for performing object and class recognition that allows for wide changes of viewpoint and distance of objects is disclosed. The invention provides for choosing pose-invariant interest points of a three-dimensional (3D) image, and for computing pose-invariant feature descriptors of the image. The system and method also allows for the construction of three-dimensional (3D) object and class models from the pose-invariant interest points and feature descriptors of previously obtained scenes. Interest points and feature descriptors of a newly acquired scene may be compared to the object and/or class models to identify the presence of an object or member of the class in the new scene.

Systems and methods are provided for extracting various features from data having spatial coordinates. The systems and methods may identify and extract data points from a point cloud, where the data points are considered to be part of the ground surface, a building, or a wire (e.g. power lines). Systems and methods are also provided for enhancing a point cloud using external data (e.g. images and other point clouds), and for tracking a moving object by comparing images with a point cloud. An objects database is also provided which can be used to scale point clouds to be of similar size. The objects database can also be used to search for certain objects in a point cloud, as well as recognize unidentified objects in a point cloud.

A method comprising: obtaining a three-dimensional (3D) point cloud about at least one object of interest; detecting ground and/or building objects from 3D point cloud data using an unsupervised segmentation method; removing the ground and/or building objects from the 3D point cloud data; and detecting one or more vertical objects from the remaining 3D point cloud data using a supervised segmentation method.

Systems and methods for cuboid detection and keypoint localization in images are disclosed. In one aspect, a deep cuboid detector can be used for simultaneous cuboid detection and keypoint localization in monocular images. The deep cuboid detector can include a plurality of convolutional layers and non-convolutional layers of a trained convolution neural network for determining a convolutional feature map from an input image. A region proposal network of the deep cuboid detector can determine a bounding box surrounding a cuboid in the image using the convolutional feature map. The pooling layer and regressor layers of the deep cuboid detector can implement iterative feature pooling for determining a refined bounding box and a parameterized representation of the cuboid.

Apparatus is arranged in operation to perform a method of estimating the number of individuals in a scene. The method comprises generating, for a plurality of image positions within at least a portion of a captured image of the scene, an edge correspondence value indicative of positional and angular correspondence with a representation of at least a partial outline of an individual. Analysis of the edge correspondence value is used to detect whether each of the plurality of image positions contributes to at least part of an image of an individual.

The present invention provides a system and method for recognizing a 3D object in a single camera image and for determining the 3D pose of the object with respect to the camera coordinate system. In one typical application, the 3D pose is used to make a robot pick up the object. A view-based approach is presented that does not show the drawbacks of previous methods because it is robust to image noise, object occlusions, clutter, and contrast changes. Furthermore, the 3D pose is determined with a high accuracy. Finally, the presented method allows the recognition of the 3D object as well as the determination of its 3D pose in a very short computation time, making it also suitable for real-time applications. These improvements are achieved by the methods disclosed herein.