Abstract: An image processing device includes a processor to detect motion of objects in an observation area on the basis of a captured image. The processor also calculates amounts of each detected motion, and calculates a propagation speed, a propagation direction, or both, of the motion for pulsations of the objects.

Abstract: A method for estimating the speed of movement of a first video camera when it captures a current image of a three-dimensional scene, includes storing a reference image corresponding to an image of the same scene captured by a second video camera in a different pose, the reference image including pixels. The method also includes storing the current image, the current image including pixels containing the measurement of a physical quantity measured by that pixel, that physical quantity being the same as the physical quantity measured by the pixels of the reference image. The method further includes storing for each pixel of the reference image or of the current image the measurement of a depth that separates that pixel from the point of the scene photographed by that pixel, estimating the pose of the first camera, and estimating the speed of movement of the first camera.

Abstract: Aspects of the disclosure generally relate to connecting panoramic images. One or more computing devices may load and display a first panoramic image captured at a first location and receive a selection of an area on the first panoramic image, the area corresponding to where a connection to other panoramic images may be made. The one or more computing devices may identify and display one or more nearby panoramic images which were captured near the first location and receive a selection of one of the one or more nearby panoramic images. The one or more computing devices may display the selected panoramic image and the first panoramic image and align the selected panoramic image with the first panoramic image such that the selected panoramic image is oriented in the same direction as the first panoramic image. The one or more computing devices may connect the selected panoramic image with the first panoramic image.

Abstract: The method may include obtaining a first set of imaging data affording a sagittal view relating to a subject and a first couch supporting the subject. The first couch may have a plurality of first positions reflected in the first set of imaging data as a first conformation. The method may also include determining a displacement field associated with a first set of imaging data with respect to the reference conformation based on the first conformation and a reference conformation. The method may further include adjusting the first set of imaging data with respect to the reference conformation based on the displacement field. In some embodiments, the method may include obtaining an image of the subject with respect to the reference conformation based on the adjusted first set of imaging data.

Abstract: A system for estimating a pose of one or more persons in a scene includes a camera configured to capture an image of the scene; and a data processor configured to execute computer executable instructions for: (i) receiving the image of the scene from the camera; (ii) extracting features from the image of the scene for providing inputs to a keypoint subnet and a person detection subnet; (iii) generating one or more keypoints using the keypoint subnet; (iv) generating one or more person instances using the person detection subnet; (v) assigning the one or more keypoints to the one or more person instances by learning pose structures from the image data; and (vi) determining one or more poses of the one or more persons in the scene using the assignment of the one or more keypoints to the one or more person instances.

Abstract: An electronic apparatus is provided. The electronic apparatus includes: a storage configured to store a plurality of filters each corresponding to a plurality of image patterns; and a processor configured to classify an image block including a target pixel and a plurality of surrounding pixels into one of the plurality of image patterns based on a relationship between pixels within the image block and to obtain a final image block in which the target pixel is image-processed by applying at least one filter corresponding to the classified image pattern from among the plurality of filters to the image block, wherein the plurality of filters are obtained by learning, through an artificial intelligence algorithm, a relationship between a plurality of first sample image blocks and a plurality of second sample image blocks corresponding to the plurality of first sample image blocks based on each of the plurality of image patterns.

Abstract: Apparatuses, systems and methods are provided for determining vehicle driver distractions. More particularly, apparatuses, systems and methods are provided for determining distractions at particular geographic locations.

Abstract: Techniques for object re-identification based on temporal context. Embodiments extract, from a first image corresponding to a first camera device and a second image corresponding to a second camera device, a first plurality of patch descriptors and a second plurality of patch descriptors, respectively. A measure of visual similarity between the first image and the second image is computed, based on the first plurality of patch descriptors and the second plurality of patch descriptors. A temporal cost between the first image and the second image is computed, based on a first timestamp at which the first image was captured and a second timestamp at which the second image was captured. The measure of visual similarity and the temporal cost are combined into a single cost function, and embodiments determine whether the first image and the second image depict a common object, using the single cost function.

Abstract: Methods, systems, and apparatuses for detecting and describing heterogeneity in a cell sample are disclosed herein. A plurality of fields of view (FOV) are generated for one or more areas of interest (AOI) within an image of the cell sample are generated. Hyperspectral or multispectral data from each FOV is organized into an image stack containing one or more z-layers, with each z-layer containing intensity data for a single marker at each pixel in the FOV. A cluster analysis is applied to the image stacks, wherein the clustering algorithm groups pixels having a similar ratio of detectable marker intensity across layers of the z-axis, thereby generating a plurality of clusters having similar expression patterns.

Abstract: A currency classification device that classifies currency types using currency images includes a feature value calculator, a storage, and an output unit. The feature value calculator calculates feature values for every currency type that is a candidate for classification from an image area common in images of every currency type. The storage stores the feature values calculated by the feature value calculator from learning images, which are currency models, as templates for every currency type. The output unit outputs the currency type corresponding to the template having a highest value of similarity with the feature value calculated by the feature value calculator an input image, which is a currency image subject to classification, in the templates stored in the storage as a classification result.

Abstract: An image processing apparatus comprises a corresponding point obtaining unit configured to obtain corresponding points that are feature points of an object associated between a first image and a second image; a first deformation obtaining unit configured to obtain first deformation that is based on the corresponding points; a second deformation obtaining unit configured to obtain second deformation that is performed between the first image and the second image, and in which a degree of coincidence of corresponding points is lower than the first deformation; and a combining unit configured to combine the first deformation and the second deformation, regarding positions in at least one image of the first Image and the second image, based on distances between the corresponding points and the positions.

Abstract: Embodiments disclosed herein provide systems and methods for determining motion of a striking or blocking device. Embodiments may be utilized in various embodiments when it is desired to track, determine, etc. the motion of a device. For example, embodiments may be utilized in martial arts, sports, activities, instructional videos, video games, etc. More specifically, embodiments may be configured to automatically detect fast movements of a device using video cameras or image capturing devices.

Abstract: Provided is an image processing device including a changing unit that changes a threshold value of intensity of each of points within an image, a first display control unit that controls displaying of points including an intensity value equal to or greater than the threshold value every time the threshold value is changed by the changing unit, a receiving unit that receives designation of one or a plurality of predetermined regions within a displayed image based on an operation of a user, a fixing unit that fixes the threshold value of each of the points within the designated predetermined region as a threshold value by the changing unit at a time of designation of the predetermined region, a second display control unit that controls displaying of points including an intensity value equal to or greater than the threshold value fixed by the fixing unit among the points within the predetermined region, and points including an intensity value equal to or greater than the threshold value to be changed by the chan

Abstract: Skeletal recording devices (e.g., Microsoft Kinect®) has been gaining popularity in home-based rehabilitation solution due to its affordability and ease of use. It is used as a marker less human skeleton tracking device. However, apart from the fact that the skeleton data are contaminated with high frequency noise, the major drawback lies in the inability to retain the anthropometric properties, for example, the body segments' length, which varies with time during the tracking. Embodiments of the present disclosure provide systems that implement a particle filter based approach to track the human skeleton data in presence of high frequency noise and multi-objective genetic technique is further implemented to reduce the bone length variations. Further multiple segments in skeleton are filtered simultaneously and segments' lengths are preserved by considering their interconnection for obtained corrected set of body joint positions which ensures that the body segment length is maintained close to ground truth.

Abstract: With regard to image data processing, a method of obtaining a focus term by using periodicity of the focus term is provided. The focus term may be used in a plurality of operation processes for processing image data.

Abstract: Embodiments of a method and system described herein enable capture of video data streams from multiple, different video data source devices and the processing of the video data streams. The video data streams are merged such that various data protocols can all be processed with the same worker processors on different types of operating systems, which are typically distributed. In an embodiment the multiple video data sources comprises at least one mobile device executing a video sensing application that produces a video data stream for processing by video analysis worker processes. The processes include automatically detecting features in an urban scene comprising building entrances.

Abstract: Body joint tracking is applied in various industries and medical field. In body joint tracking, marker less devices plays an important role. However, the marker less devices are facing some challenges in providing optimal tracking due to occlusion, ambiguity, lighting conditions, dynamic objects etc. System and method of the present disclosure provides an optimized body joint tracking. Here, motion data pertaining to a first set of motion frames from a motion sensor are received. Further, the motion data are processed to obtain a plurality of 3 dimensional cylindrical models. Here, every cylindrical model among the plurality of 3 dimensional cylindrical model represents a body segment. The coefficients associated with the plurality of 3 dimensional cylindrical models are initialized to obtain a set of initialized cylindrical models. A set of dynamic coefficients associated with the initialized cylindrical models are utilized to track joint motion trajectories of a set of subsequent frames.

Abstract: A system includes an aerial sensor platform including a spectral imaging device, a position sensor, and an orientation sensor and includes a ground-based sensor platform including at least one soil sensor. The system also includes a computing device with instructions that are executable by a processor to obtain spectral imaging data collected by the spectral imaging device and soil data collected by the at least one soil sensor. The spectral imaging data represents a particular field of view (based on data from the orientation sensor) of a particular geographic region (based on data from the position sensor) of a crop field and the at least one soil sensor is associated with the particular geographic region. The instructions are further executable by the processor to schedule an agricultural activity based on the spectral imaging data and the soil data.

Abstract: A method and a device for measuring the topography and/or the gradients and/or the curvature of an optically active surface of an object are disclosed. The device allows the object to be arranged in a receiving region with a contact surface for contact with the object. Inside the device, there is a plurality of point light sources that provide light that is reflected at the surface to be measured of an object arranged in the receiving region. The device includes at least one camera with an objective assembly and an image sensor for detecting a brightness distribution which is produced on a light sensor by the light of the point light sources reflected at the surface to be measured.

Abstract: A lane detection method based on vision and lidar multi-level fusion includes: calibrating obtained point cloud data and an obtained video image; constructing a point cloud clustering model by fusing height information, reflection intensity information of the point cloud data, and RGB information of the video image, obtaining point clouds of a road based on the point cloud clustering model, and obtaining a lane surface as a first lane candidate region by performing least square fitting on the point clouds; obtaining four-channel road information by fusing the reflection intensity information of the point cloud data and the RGB information of the video image, inputting the four-channel road information into the semantic segmentation network 3D-LaneNet, and outputting an image of a second lane candidate region; and fusing the first lane candidate region and the second lane candidate region, and combining the two lane candidate regions into a final lane region.