Abstract: Methods and systems for determining and classifying a number of repetitive motions in a video are described, and include the steps of first determining a plurality of images from a video, where the images are segmented from at least one video frame of the video. Next, performing a pose detection process on a feature of the images to generate one or more landmarks. Next, determining one or more principle component axes on points associated with a given landmark. Finally, determining at least one repetitive motion based on a pattern associated with a projection of the points onto the one or more principle components. In some embodiments, the disclosed methods can classify the repetitive motions to respective types. The present invention can be implemented for convenient use on a mobile computing device, such as a smartphone, for tracking exercises and similar repetitive motions.

Abstract: An unmarked body motion tracking system based on multi-Kinect V2 is provided and includes: Kinect V2s, configured to acquire depth data and joint data; analysis modules, configured to analyze data collection characteristics of single Kinect V2 and transmission device assembly requirements; clients, configured to receive and process data collected by Kinect V2s. A process of the data processing includes: based on the data collection characteristics of the single Kinect V2, a client-server model is built to track a human body from different directions; output modules, configured to output layout modes and tracking results of the Kinect V2s. According to the data collection characteristics of the single Kinect V2 and the transmission device assembly requirements, the disclosure minimizes the mutual interference between the opposite Kinect V2s, tracks the human body from different directions, and ensures that the joints in the self-occlusion state are stably tracked by sensors at other angles.

Abstract: Embodiments of the present disclosure disclose an anomaly detector for detecting an anomaly in a sequence of poses of a human performing an activity. The anomaly detector includes an input interface configured to accept input data indicative of a distribution of the sequence of poses, a memory configured to store a discriminative one-class classifier having a pair of complementary classifiers bounding normal distribution of pose sequences in a reproducing kernel Hilbert space (RKHS), a processor configured to embed the input data into an element of the RKHS and classify the embedded data using the discriminative one-class classifier, and an output interface configured to render a classification result.

Abstract: Aspects of the present disclosure include methods, systems, and non-transitory computer readable media that perform the steps of receiving a first plurality of snapshots, generating a first plurality of descriptors each associated with the first plurality of snapshots, grouping the first plurality of snapshots into at least one cluster based on the plurality of descriptors, selecting a representative snapshot for each of the at least one cluster, generating at least one second descriptor for the representative snapshot for each of the at least one cluster, wherein the at least one second descriptor is more complex than the first plurality of descriptors, and identifying a target by applying the at least second descriptor to a second plurality of snapshots.

Abstract: A method for tracking movement of a presenter during a videoconference comprises tracking the location of subject of interest; quickly detecting error conditions—such as can occur when a subject of interest moves too quickly or when a subject of interest walks past another person—and quickly reinitiating the tracking sequence to help ensure a positive videoconferencing experience.

Abstract: A system for identification of bacterial cells in free solution in a sample. A sample handler is adapted to position the sample. A light source illuminates a large volume of the sample. An imager is located to receive light scattered from the sample. A computer it is coupled to receive data transmitted from the imager. A controller is coupled to send control signals to the sample handler and the computer. The imager processes the scattered light to form images of the bacteria and transmits bacteria image information to the computer, wherein the bacteria image information includes intensity values and position data for the bacteria images from which the computer determines the presence of bacteria.

Abstract: An information processing apparatus constructs a virtual space to express a subject by acquiring an image obtained by observing a reality space in which one or more subjects are present, and arranging a plurality of unit volume elements at positions that are determined depending on a subject captured in the acquired image, and the information processing apparatus constructs a virtual space by arranging multiple types of unit volume elements with different sizes.

Abstract: An image analysis device includes a person detecting unit for detecting person images from image information from a surveillance camera, and acquiring both pieces of person image information as a result of the detection of person images, and pieces of attribution information about persons corresponding to the pieces of person image information, and a person narrowing down unit for performing narrowing-down of person image information with respect to the pieces of person image information and the pieces of person attribution information which are acquired by the person detecting unit, on the basis of a person narrowing down rule, to acquire person image information after the narrowing-down and attribution information about a person corresponding to the person image information.

Abstract: An obstacle recognition device of a vehicle provided with a camera capturing an image around the vehicle, includes an acquiring unit sequentially acquiring the image captured by the camera; a feature point extracting unit extracting a plurality of feature points of an object included in the image; a calculation unit calculating each motion distance of the plurality of feature points between the image previously acquired and the image currently acquired by the acquiring unit; a first determination unit determining whether each motion distance of the feature points is larger than or equal to a first threshold; a second determination unit determining whether each motion distance of the feature points is larger than or equal to a second threshold; and an obstacle recognition unit recognizing an obstacle.

Abstract: A method and an apparatus for recognizing a descriptive attribute of an appearance feature include obtaining a location feature of an appearance feature of a target image to determine a location of a part of an object in a preset object model indicated by the appearance feature, where the location feature of the appearance feature indicates the location of the part of the object in the preset object model indicated by the appearance feature, recognizing a target region based on the location feature, where the target region includes the part of the object, performing feature analysis on the target region, recognizing a descriptive attribute of the appearance feature of the object, and determining the location feature of the appearance feature having a local attribute.

Abstract: A method, system and computer readable storage media for segmenting individual intra-oral measurements and registering said individual intraoral measurements to eliminate or reduce registration errors. An operator may use a dental camera to scan teeth and a trained deep neural network may automatically detect portions of the input images that can cause registration errors and reduce or eliminate the effect of these sources of registration errors.

Abstract: Provided are a method and an apparatus wherein an object is distinguished from the background of the object by using a sensor unit including at least one sensor, the distances to measurement points of the background and/or the object are measured, and information on the shape of at least one part constituting the object is acquired on the basis of the measured distances.

Abstract: An operating assistance method for a working device or for a vehicle. Object boxes for an object in a field of view of the working device are obtained at consecutive times. From object boxes of a given object, for images recorded in succession or direct succession, an instantaneous change of scale of an object box for the specific object, and an instantaneous lateral change of position of the object box for the specific object, are determined. An object box predicted in the future is determined from the current change of scale, and from the instantaneous lateral change in position for an object box for a specific object. The position of the predicted object box and/or the ratio of a lateral extension of the predicted object box to a lateral extension of a covered field of view and/or of the recorded images are determined and evaluated.

Abstract: To identify and track a plurality of objects. An object tracking device 1 includes a frame acquisition unit 31 for acquiring one set of frame data 11 included in moving image data that captures a space where a plurality of objects are present; a distance acquisition unit 32 for acquiring measurement data 12 from measurement of three-dimensional coordinates of points configuring the objects in the space where the objects are present; an object position detection unit 33 for detecting a two-dimensional position of each of the objects from the frame data 11 on the basis of comparison with template data of each of the objects; an object distance detection unit 34 for detecting a plurality of three-dimensional positions of the objects from the measurement data 12; and an associating unit 35 for associating each two-dimensional position from the object position detection unit 33 with the nearest three-dimensional position among the three-dimensional positions from the object distance detection unit 34.

Abstract: A multi-task hierarchical machine learning model is configured to perform both a decision task to predict a decision result and an explanation task to predict a plurality of semantic concepts for explainability associated with the decision task, wherein a semantic layer of the machine learning model associated with the explanation task is utilized as an input to a subsequent decision layer of the machine learning model associated with the decision task. Training data is received. The multi-task hierarchical machine learning model is trained using the received training data.

Abstract: A particle velocimetry system and algorithm are provided for executing a particle reconstruction to determine three-dimensional positions of particles in a particle laden fluid flow from two-dimensional flow images generated by two-dimensional image sensors; generate a three-dimensional particle distribution from the three-dimensional position; and execute a recursive loop for performing further iterations of particle reconstruction and generation of three-dimensional particle distributions, with recursive iterations of particle reconstruction executed with the use of particle property data obtained from the prior executed iteration of particle reconstruction.

Abstract: A method for identifying a risk merchant comprises: determining, for each of at least one marked merchant cluster based on a user correlation degree of a merchant to be identified and the each marked merchant cluster, a degree of similarity between the merchant to be identified and the marked merchant cluster, wherein each marked merchant in the each marked merchant cluster is marked with a corresponding risk category by a trained classification model, and the user correlation degree of the merchant to be identified and the marked merchant cluster comprises a user correlation coefficient of the merchant to be identified and the marked merchant cluster and a number of same users of the merchant to be identified and the marked merchant cluster; and determining whether the merchant to be identified is a risk merchant based on the degree of similarity between the merchant to be identified and the each marked merchant cluster.

Abstract: The present disclosure provides a method for detecting illegal parking. The method includes: obtaining a detection frame having a category of vehicle from an image acquired by an acquisition module through a first neural network to determine coordinate information of the detection frame; determining whether an area corresponding to the detection frame in the image is an illegal parking area through a second neural network; and when determining that the area is the illegal parking area, determining whether the coordinate information of the detection frame is changed within a preset period of time to determine whether the vehicle is parked illegally.

Abstract: A checking device for a label, with a processor for the detection of the label and a layout unit for the creation of a layout of the label from layout data, wherein the checking device comprises a data converter device configured in such a way that it generates test data from layout data in memory in order to control the processor. A data converter device for use in a checking device, wherein the data converter device is configured in such a way that it generates test data from layout data in memory for checking the label.

Abstract: Systems and methods for estimating a height of an object from a monocular image are described herein. Objects are detected in the image, each object being indicated by a region of interest. The image is then cropped for each region of interest and the cropped image scaled to a predetermined size. The cropped and scaled image is then input into a convolutional neural network (CNN), the output of which is an estimated height for the object. The height may be represented by a mean of a probability distribution of possible sizes, a standard deviation, as well as a level of confidence. A location of the object may be determined based on the estimated height and region of interest. A ground truth dataset may be generated for training the CNN by simultaneously capturing a LIDAR sequence with a monocular image sequence.