Abstract: Methods, systems and processor-readable media for classifying human coactivity performed jointly by two humans shown in an image or a sequence of frames of a video. A 2D convolutional neural network is used to identify key points on the human body, such as human body joints, visible within the image or within each frame, for each of the two people performing the coactivity. An encoded representation of the key points is created for each image or frame, the encoded representation being based on distances between the key points of the first person and key points of the second person. The encoded representation for the image, or a concatenated volume of the encoded representations of the frames, is processed by a fully-connected neural network trained to classify the coactivity.

Abstract: A taxis analysis method for performing taxis analysis of nematodes using a container in which a reference point is provided in the container or a culture medium in the container, the taxis analysis method including steps of: imaging a distribution mode of the nematodes in the container after the nematodes and a specimen of a subject are dropped into the container; detecting a position of an object of the reference point corresponding to the reference point included in the image obtained by imaging; determining an attraction region and a avoidance region on the basis of the position; and executing taxis analysis using objects of the nematodes in the determined attraction region and avoidance region.

Abstract: Methods and systems of frame based image segmentation are provided. For example, a method for feature object tracking between frames of video data is provided. The method comprises receiving a first frame of video data, extracting a mask feature for each of one or more objects of the first frame, adjusting the first frame by applying each initial mask and corresponding identification to a respective object of the first frame, and outputting the adjusted first frame. The method further comprises tracking the one or more objects in one or more consecutive frames. The tracking comprises extracting a masked feature for each of one or more objects in the consecutive frame, adjusting the consecutive frame by applying each initial mask and corresponding identification for the consecutive frame to the respective object of the one or more objects of the consecutive frame, and outputting the adjusted consecutive frame.

Abstract: Provided is a method and apparatus for detecting a planar surface, the method including acquiring, based on a pixelwise disparity of an input image estimated in a first network, a pixelwise plane parameter of the input image, determining a pixelwise segment matching probability of the input image based on a second network trained to perform a segmentation of an image, acquiring a segment-wise plane parameter based on the pixelwise plane parameter and the pixelwise segment matching probability, and detecting a planar surface in the input image based on the segment-wise plane parameter.

Abstract: The information processing device includes a processor. The processor performs: a reconstruction process of reconstructing a plurality of tomographic images from a plurality of projection images obtained by tomosynthesis imaging that irradiates an object with radiation from a radiation source at a plurality of radiation source positions; a synthesis process of synthesizing the plurality of tomographic images to generate a synthesized two-dimensional image as a projection image based on a virtual radiation source position; an anatomical structure extraction process of extracting anatomical structures from a projection image corresponding to a radiation source position closest to the virtual radiation source position among the plurality of radiation source positions; and a display process of displaying an extraction result of the anatomical structures on a display device together with the synthesized two-dimensional image.

Abstract: Methods and systems for detecting dynamic objects using a mesh network of nodes coupled to cameras are disclosed. The system can receive sequences of frames captured by a first capture device having a first pose and a second capture device having a second pose, and track objects over time across the first sequence of frames and the second sequence of frames. The system can map the objects to three-dimensional (3D) positions in a 3D coordinate space based on correspondences between the indications of the objects. The system can determine a 3D displacement of a subset of the 3D points, and generate a 3D volume surrounding the subset. The system can use the 3D volume to classify and predict a trajectory for the dynamic object, as well as determine a risk the dynamic object poses to a protected volume in the 3D coordinate space.

Abstract: A method of generating a training dataset suitable for training machine learning algorithms to estimate the motion of objects, and for training a machine learning algorithm to perform motion estimation. A plurality of pairs of synthetic images are generated from obtained objects and backgrounds, each pair have a first frame and a second frame. The first frame includes a selection of objects in first positions and first orientations superimposed on a selected background, and the second frame includes the selection of objects in second positions and second orientations superimposed on the selected background. Also provided are processing systems configured to carry out these methods.

Abstract: A visual positioning method and apparatus are provided. In some embodiments, the method includes: acquiring a video captured by an image sensor; determining visual positioning information respectively corresponding to a plurality of key image frames in the video; determining a capture pose transformation relationship between each of the plurality of key image frames according to inertial navigation information of the image sensor recorded when taking the video; performing, according to the visual positioning information corresponding to each of the plurality of key image frames, graph optimization processing on the visual positioning information corresponding to each of the plurality of key image frames by using the capture pose transformation relationship between each of the plurality of key image frames as an edge constraint; and determining, according to a result of the graph optimization processing, a visual positioning result of the image sensor when taking the video.

Abstract: An illustrative, non-limiting method for resolving Region-On-Interest (ROI) overlaps includes receiving Simultaneous Localization and Mapping (SLAM) data obtained by a plurality of co-located Head-Mounted Devices (HMD), where the SLAM data comprises a plurality of landmarks and each HMD has an Information Handling System (IHS) in communication therewith. Each IHS may produce a subset of the plurality of landmarks based upon a respective HMD's SLAM data, and each IHS is: local with respect to the plurality of HMDs, at an edge of a network serving an HMD, or on a cloud. The method may include identifying an overlap between a first landmark produced by a first IHS and a second landmark produced by a second IHS, selecting between the first and second landmarks, at least in part, based upon the locations of the first and second IHSs, and rendering a map for display by a given HMD using the selected landmark.

Abstract: Provided are a system, method, and computer readable storage medium for generating an intra-oral scan of a dentition. A contiguous scan of an occlusal surface of at least a part of one arch of the dentition is generated. At least one additional surface is associated to the contiguous scan of the occlusal surface.

Abstract: An information processing apparatus comprises: a first obtaining unit configured to obtain feature vectors related to a first category from a vector group; a second obtaining unit configured to obtain feature vectors related to a second category from the vector group; a calculation unit configured to calculate a transformation matrix for converting a projection source vector obtained based on a feature vector obtained by the second obtaining unit into a projection destination vector obtained based on a feature vector obtained by the first obtaining unit; and a generation unit configured to generate a pseudo feature vector related to a third category by performing conversion processing on feature vectors included in the feature vectors obtained by the second obtaining unit using the transformation matrix.

Abstract: A method and workstation for combining multiple functional imaging datasets into a joint-visualization image. In one aspect, a method of functional imaging includes accessing a plurality of functional imaging datasets acquired from a volume-of-interest, wherein each of the plurality of functional imaging datasets is a different one of a plurality of functional imaging data types. The method includes registering the plurality of functional imaging datasets and determining a visualization priority for each of a plurality of pixels in a joint-visualization image based on a logical comparison of corresponding information in each of the plurality of functional imaging datasets. The method includes generating the joint-visualization image based on the visualization priority and at least a portion of each of the plurality of functional imaging datasets, wherein each of a plurality of pixels in the joint-visualization image represents only a single one of the plurality of functional imaging data types.

Abstract: A method for calculating a distance between a vehicle camera and an object, the method may include: (a) obtaining an image that was acquired by the vehicle camera of a vehicle; the image captures the horizon, the object, and road lane boundaries; (b) determining an initial row-location horizon estimate and a row-location contact point estimate, the contact point is between the object and a road on which the vehicle is positioned; (c) determining a vehicle camera roll angle correction that once applied will cause the lanes boundaries to be parallel to each other in the real world; (d) calculating a new row-location horizon estimate, wherein the calculating comprises updating the row-location horizon estimate based on the vehicle camera roll angle correction; and (e) calculating the distance between the vehicle camera based on a difference between the new row-location horizon estimate and the row-location contact point estimate.

Abstract: A computing system determines accuracy of sport-related information extracted from a time sequence of digital video frames that represent a sport event, the extracted sport-related information including an attribute that changes over the time sequence. The computing system (a) detects, based on the extracted sport-related information, a pattern of change of the attribute over the time sequence and (b) makes a determination of whether the detected pattern is an expected pattern of change associated with the sport event. If the determination is that the detected pattern is the expected pattern, then, responsive to making the determination, the computing system takes a first action that corresponds to the sport-related information being accurate. Whereas, if the determination is that the detected pattern is not the expected pattern, then, responsive to making the determination, the computing system takes a second action that corresponds to the sport-related information being inaccurate.

Abstract: A method for classifying and counting objects recoverable from a urine sample processed onto a slide. The method includes the following steps: receiving at least one digitalized image of the whole slide; detecting connected components by segmentation of the image of the whole slide; classifying the detected connected components into countable connected components and uncountable connected components using a classifier; for the countable connected components using an object detection model to obtaining the number of objects for each class; for the uncountable components using a semantic segmentation model to obtaining the number of objects for each class; summing up the number of objects for each class obtained from the semantic segmentation model and the object detection model outputting a number of objects for each class.

Abstract: The present invention relates to the field of radiation therapy and methods, software and systems for image registration. According to the present invention, methods for image registration of images of a patient comprising are provided. The method may include receiving a set of image data from the patient; selecting one or more tracking regions in the set of image data, each tracking region including a tracking point or set of tracking points; receiving sets of image data captured at different times from the patient; identifying corresponding tracking regions in the sets of image data captured at different times; estimating a displacement field describing displacement vectors for a number of points in the sets of image data including the tracking regions, each displacement vector specifying a position of a region in reference to a previous position.

Abstract: Upon a triggering event, frames of a video feed are analyzed for activity or motion. One frame or a subset of the frames are selected as a keyframe/keyframes based on one or measures associated with one or more metrics derived from the frames. The keyframe/keyframes optimally reflects an activity or degree of motion across the frames. The keyframe/keyframes are then provided as input to a computer vision or machine learning application to improve the efficiency, accuracy, and/or precision of results produced by or decisions made by the application.

Abstract: The present disclosure relates to systems, non-transitory computer-readable media, and methods for adapting generative neural networks to target domains utilizing an image translation neural network. In particular, in one or more embodiments, the disclosed systems utilize an image translation neural network to translate target results to a source domain for input in target neural network adaptation. For instance, in some embodiments, the disclosed systems compare a translated target result with a source result from a pretrained source generative neural network to adjust parameters of a target generative neural network to produce results corresponding in features to source results and corresponding in style to the target domain.

Abstract: An image processing method for a video processor, for generating an extrapolated frame according to a previous frame and a current frame, includes steps of: projecting a plurality of motion vectors (MVs) to the extrapolated frame subsequent to the current frame; determining whether a block of the extrapolated frame is projected by at least two of the MVs; selecting at least two candidate MVs from the MVs projected to the block when the block is projected by at least two of the MVs; calculating a blended MV which is a mixture of the at least two candidate MVs, and projecting the blended MV to the previous frame; obtaining a reference MV corresponding to position of the previous frame projected by the blended MV; and comparing the reference MV with the at least two candidate MVs, to select a final MV for the block from the at least two candidate MVs.

Abstract: An evaluation method and system for the corrosion degree of an absorbable stent.