Abstract: A method, a system, and a non-transitory computer readable medium for evaluating first movements of a first person in view of movements of a second person. The method may include determining the relationships between the first frames and the second frames by a sequence alignment algorithm, the determining of the relationships is based on the inter-frame similarity scores and the multiple state-related scores; and determining at least one evaluation score of the movements of the first person, based on the relationships and the multiple inter-frame similarity scores; wherein the at least one evaluation score is indicative of a similarity between the movements of the first person and the movements of the second person.

Abstract: One or more computing devices perform a method for placing a virtual tripwire onto grid points of a geocoordinate grid. A device processor directs a camera to capture a series of images within the camera's field of view. The processor identifies objects of interest within the images and determines movements of the objects through the images. The processor identifies directions of travel of the objects and determines variances in those directions of travel. The processor further determines vectors of median direction of travel for vector points corresponding to the objects on the grid. The processor then determines multiple candidate locations for placement of the tripwire across grid points of the grid in a direction approximately orthogonal to the direction of travel of at least one object of interest and automatically selects one of the candidate locations based on the vectors of median direction of travel and the travel direction variances.

Abstract: A method and an active commissioning system for video analytics (VA) of video surveillance are provided. The method includes capturing, using a video capture device, a video sequence and receiving, at a VA processor, a parameter set from a cloud resource, wherein the parameter set is used for video analytics. The method also includes processing the video sequence by selecting subparts of the video sequence as events, wherein the events are detected based on the parameter set, generating a visualized report using the events from the VA processor and providing the visualized report to an operator, marking, by the operator, yes/no on the events in the visualized report, generating annotations for events based on the marking, and transmitting the annotations and the events to the cloud resource for processing, wherein the cloud resource processes the annotations and events and provides the parameter set based on the processing.

Abstract: A behavior recognition method includes acquiring positions of a plurality of respective joints and recognizing a partial behavior that is a subject to be recognized based on a feature value that is calculated from the positions of the respective joints using weights that are to be applied to the respective joints and that are changed according to a type of the partial behavior, by a processor.

Abstract: A method for detecting a static logo of a video, an electronic apparatus and a storage medium. The method includes: calculating a pixel grayscale flag value, an edge gradient flag value and an edge direction flag value; calculating, in a preset neighborhood centered on the pixel at each pixel position of the current video frame, a first local confidence degree of the pixel grayscale flag value, a second local confidence degree of the edge gradient flag value, and a third local confidence degree of the edge direction flag value respectively; calculating a contribution score of each local confidence degree and a total contribution score of each pixel position; and gathering the total contribution score of each pixel position of the current video frame, and determining a static logo in the current video frame according to the total contribution score of each pixel position.

Abstract: A method for monocular depth/pose estimation in a camera agnostic network is described. The method includes projecting lifted 3D points onto an image plane according to a predicted ray vector based on a monocular depth model, a monocular pose model, and a camera center of a camera agnostic network. The method also includes predicting a warped target image from a predicted depth map of the monocular depth model, a ray surface of the predicted ray vector, and a projection of the lifted 3D points according to the camera agnostic network.

Abstract: The present invention relates to a device (10) for detecting a misuse of a medical imaging system (20), comprising a data interface (12) for acquiring medical image data (24) and audit log data (26) from the medical imaging system (20); a processing unit (14) which is configured to configured to analyse the medical image data (24) to determine whether or not a part of a fetus is imaged in the medical image data (24), to compare the medical image data (24) and the audit log data (26) with each other, and to determine based on said comparison whether there is a mismatch between the medical image data (24) and the audit log data (26); and a feedback unit (16) which is configured to generate a misuse alert signal if a mismatch is detected by the processing unit (14).

Abstract: Described are systems and methods for determining an agent that performed an event within a materials handling facility. A series of overhead images that include representations of the event location and one or more agents are processed to determine a motion or movement of the agent over a period of time. For example, a motion model representative of a motion of the agent over a period of time is generated from the images. A distance between the motion model and the event location is also determined. An association between the agent and the event may be determined based on the motion model and the distance between the motion model and the event location.

Abstract: Techniques for inferring whether an event is occurring in 3D space based on 2D image data and for maintaining a camera's calibration are disclosed. An image of an environment is accessed. Input is received, where the input includes a 2D rule imposed against a ground plane. The 2D rule includes conditions indicative of an event. A bounding box is generated and encompasses a detected object. A point within the bounding box is projected from a 2D-space image plane of the image into 3D space to generate a 3D-space point. Based on the 3D-space point, a 3D-space ground contact point is generated. That 3D-space ground contact point is reprojected onto the ground plane of the image to generate a synthesized 2D ground contact point. A location of the synthesized 2D ground contact point is determined to satisfy the conditions.

Abstract: A system and method for space management monitoring and reporting using video analytics are disclosed. Such a building occupancy monitoring system monitors building resources that include rooms, areas such as work areas, and chairs and desks in the rooms, in examples. These building resources are monitored to determine whether the resources are being used by individuals, and for how long. In more detail, the system includes an image analytics module that tracks individuals in image data from surveillance cameras, relative to the building resources. The system also includes an occupancy report module that generates occupancy information including building resources that are used/unused by the individuals. The occupancy information is then used to program building control systems. Dwell maps and motion maps generated by the system, based on the occupancy information, can also be used to guide individuals to available building resources in a ‘hot desking’ environment.

Abstract: The present disclosure involves a reference line determination method and system. In a process of determining a reference line, a plurality of original images containing a first spatial position information are obtained. According to the plurality of original images, a composite image containing a second spatial position information is further determined. After a composition relationship between a plurality of original images was determined, a reference line is determined on the composite image according to the spatial position information.

Abstract: A system for performing synergistic object tracking and pattern recognition for event representation includes a computing platform having processing hardware and a system memory storing a software code. The processing hardware is configured to execute the software code to receive event data corresponding to one or more propertie(s) of an object, to generate, using the event data, a location data estimating a location of each of multiple predetermined landmarks of the object, and to predict, using one or both of the event data and the location data, a pattern corresponding to the propertie(s) of the object. The processing hardware is further configured to execute the software code to update, using the predicted pattern, the location data, and to merge the updated location data and the predicted pattern to provide merged data.

Abstract: An image processing apparatus detects a tracking target object in an image, executes tracking processing to track the object, determines whether an attribute of an object detected from the image is a predetermined attribute, identifies, when a first state in which the object is detected changes to a second state with the object not detected, a given object included in the image and positioned at least partially in front of the object in the second state, based on a position of the object in the first state, controls the tracking processing, based on a determination whether an attribute of the given object is the predetermined attribute, and determines, based on the result, whether to continue the tracking processing. The tracking processing continues until at least a predetermined time has elapsed when a determination to continue the tracking processing on the tracking target object for the predetermined time has been made.

Abstract: A closed-loop detecting method using an inverted index-based key frame selection strategy, storage media and apparatus are provided. The method includes following steps: step I: acquiring image information at a current position, processing the image information to extract corresponding image features therefrom and solve a camera pose; step II: capturing image features successively during movement of a robot, as consecutive image frames, performing, on the consecutive image frames, an indexing, in which the inverted index-based key frame selection strategy is introduced into a key frame selection strategy to supplement key frames which are prone to be missed in a conventional forward indexing during a curvilinear movement of the robot; and step III: performing closed-loop detection and correction of accumulative errors based on image features carried by key frames.

Abstract: A robotic system includes a processor that is programmed to determine and cause work site measurements for user specified points in the work site to be graphically displayed in order to provide geometrically appropriate tool selection assistance to the user. The processor is also programmed to determine an optimal one of a plurality of tools of varying geometries for use at the work site and to cause graphical representations of at least the optimal tool to be displayed along with the work site measurements.

Abstract: Methods and products for automated real-time ovarian follicular detection, monitoring and analysis are provided. The devices and methods allow for remote or local analysis, while minimizing or eliminating the need for technician review of the output images. The methods are useful in human and non-human subjects including companion animals and other animals such as endangered species.

Abstract: A system includes a sensor and a tracking subsystem. The tracking subsystem receives an image feed of top-view images generated by the sensor and detects an event associated with a portion of a person entering a zone adjacent to a rack. The tracking subsystem determines that a first and second person may be associated with the event. The subsystem tracks the item and calculates a velocity of the item as it is moved through the space. The subsystem identifies, based on the calculated velocity, a frame in which the velocity of the item is less than a threshold velocity. The subsystem determines whether the first or second person is nearer the item in the identified frame. If the first person is nearer, the item is assigned to the first person.

Abstract: A generative adversarial network (GAN)-based system for generating a color image from an edge image includes: a first GAN training a model for converting the edge image into an intermediate image; and a second GAN training a model for converting the intermediate image into the color image, wherein an entropy of the intermediate image corresponds to a value between an entropy of the edge image and an entropy of the color image.

Abstract: A method of removing raindrops from video images is provided. The method includes the steps of: training a raindrop image recognition model using a plurality raindrop training images labeled in a plurality of rainy-scene images; recognizing a plurality of raindrop images from a plurality of scene images in a video sequence using the raindrop image recognition model; and in response to a specific raindrop image in a current scene image satisfying a predetermined condition, replacing the specific raindrop image in the current scene image with an image region corresponding to the specific raindrop image in a specific scene image prior to the current scene image to generate an output scene image.

Abstract: Systems and methods for depth estimation in accordance with embodiments of the invention are illustrated. One embodiment includes a method for estimating depth from images. The method includes steps for receiving a plurality of source images captured from a plurality of different viewpoints using a processing system configured by an image processing application, generating a target image from a target viewpoint that is different to the viewpoints of the plurality of source images based upon a set of generative model parameters using the processing system configured by the image processing application, and identifying depth information of at least one output image based on the predicted target image using the processing system configured by the image processing application.