Abstract: Occlusion of facial features may be detected and assessed in an image captured by a camera on a device. Landmark heat maps may be used to estimate the location of landmarks such as the eyes, mouth, and nose of a user's face in the captured image. An occlusion heat map may also be generated for the captured image. The occlusion heat map may include values representing the amount of occlusion in regions of the face. The estimated locations of the eyes, mouth, and nose may be used in combination with the occlusion heat map to assess occlusion scores for the landmarks. The occlusion scores for the landmarks may be used control one or more operations of the device.

Abstract: Accuracy in interference determination between a hand gripping a workpiece and nearby objects is increased. An information processing apparatus includes a measuring unit configured to decide an object to be gripped among a plurality of objects on the basis of a first image of the imaged objects, a specifying unit configured to specify an attention area for determining, when a gripping device grips the object to be gripped, whether the gripping device interferes with objects near the object to be gripped, a controller configured to change an imaging range of an imaging device on the basis of the attention area, and a determination unit configured to determine, when the gripping device grips the object to be gripped, whether the gripping device interferes with the objects near the object to be gripped on the basis of a second image of an object imaged in a changed imaging range.

Abstract: A computer-implemented method for automatically detecting a target object from a 3D image is disclosed. The method may include receiving the 3D image acquired by an imaging device. The method may further include detecting, by a processor, a plurality of bounding boxes as containing the target object using a 3D learning network. The learning network may be trained to generate a plurality of feature maps of varying scales based on the 3D image. The method may also include determining, by the processor, a set of parameters identifying each detected bounding box using the 3D learning network, and locating, by the processor, the target object based on the set of parameters.

Abstract: An object recognition device according to an embodiment includes a camera that captures an image of an imaging area. A storage device stores, for each of a plurality of registered objects, dictionary feature information for identifying the corresponding object and dictionary boundary information for identifying an actual boundary area of the corresponding object. A processor receives the captured image from the camera, and determines an object area in the captured image. The processor extracts feature information from the object area, and, based on the extracted feature information compared to the dictionary feature information, identifies each object included in the object area. The processor also extracts boundary information corresponding to each identified object included in the object area, and, based on the extracted boundary information compared to the dictionary boundary information with respect to each identified object, determines an overlap state of each identified object in the object area.

Abstract: System, processes and devices for monitoring betting activities using bet recognition devices and a server. Each bet recognition device has an imaging component for capturing image data for a gaming table surface. The bet recognition device receives calibration data for calibrating the bet recognition device. A server processor coupled to a data store processes the image data received from the bet recognition devices over the network to detect, for each betting area, a number of chips and a final bet value for the chips.

Abstract: An imaging portion acquires first sample-images of a first sample under optical-conditions, the first sample having a defect, and acquires second sample-images of a second sample under optical-conditions, the second sample having no defects. An arithmetic portion calculates a first difference between a first sample-image taken under a first optical-condition and the first sample-image taken under a second optical-condition, calculates a second difference between a second sample-image taken under the first optical-condition and a second sample-image taken under the second optical-condition, and selects the first and second optical-conditions under which a difference between the first and second differences becomes largest, as a first and a second inspection-condition. The imaging portion takes images of the target to be inspected under the first and second inspection-conditions to acquire a first and second inspection-images.

Abstract: An information processing apparatus predicts a waiting time by suppressing a reduction in accuracy of waiting-time prediction even if the number of people waiting in a line is small. The information processing apparatus includes a calculation unit that calculates a waiting time, based on a first set of counted objects in a case where the number of first set of counted objects is less than a predetermined number and based on a second set of counted objects in a case where the first set of counted objects is greater than or equal to the predetermined number.

Abstract: An image capturing device includes an image capturing unit, a light filter module, a light sensing unit, an infrared light source and a processing unit. The light filter module includes a first light filter and a second light filter. The first light filter filters an infrared and allows a visible light to pass. The second light filter filters a visible light and allows an infrared to pass. The light sensing unit senses an environmental light. When the processing unit determines that an intensity of the environmental light is higher than a threshold, the processing unit alternately switches the first light filter and the second light filter to a front of the image capturing unit, controls the image capturing unit to capture a visible light image through the first light filter, and controls the image capturing unit to capture an infrared image through the second light filter.

Abstract: In a method and magnetic resonance (MR) apparatus for echo-planar MR imaging with which MR signals are entered in raw-data space with a zigzag-type trajectory, a sequence of readout gradients and phase-encoding gradients is applied such that a zigzag-type undersampled trajectory in raw-data space is filled, such that with the signal echoes being shifted by up to a quarter of the acquired raw-data space in the direction of the readout axis. Image data are reconstructed using a parallel-imaging reconstruction algorithm that operates on the raw data acquired in the zigzag-type trajectory, based on an interlaced Fourier transform.

Abstract: Described is a system for predicting multi-agent movements. A Radon Cumulative Distribution Transform (Radon-CDT) is applied to pairs of signature-formations representing agent movements. Canonical correlation analysis (CCA) components are identified for the pairs of signature-formations. Then, a relationship between the pairs of signature formations is learned using the CCA components. A counter signature-formation for a new dataset is predicted using the learned relationship and a new signature-formation. Control parameters of a device can be adjusted based on the predicted counter signature-formation.

Abstract: An apparatus for scaling images is provided that includes at least two input ports, a scaling component coupled to the at least two input ports, the scaling component including a plurality of scalers, the scaling component configurable to map any scaler to any input port of the at least two input ports and configurable to map more than one scaler to any input port, and a memory coupled to the at least two input ports and to outputs of the plurality of scalers, the memory configured to store image data for each input port and scaled image data output by the plurality of scalers.

Abstract: A convolutional neural network-based image processing method is provided. The method includes: receiving, in a second layer, multi-channel feature map images generated by applying a convolution operation to an input image of a convolutional neural network having a plurality of layers with a plurality of filter kernels of a first layer; analyzing a dynamic range of the multi-channel feature map images; re-ordering the multi-channel feature map images, based on the dynamic range; and processing the re-ordered multi-channel feature map images in the second layer.

Abstract: Current technologies detect the alertness of a user using frontal face. However, there are no techniques to track the position of face from a side angle. A method and system for face position tracking of a user and alerting the user is disclosed. The method includes capturing images using a monocular camera and identifying a user's face from the images and storing the user's face as reference face. The user's face is divided into two regions and one or more corner points are identified. A centroid is created in the two regions and the corner points are joined using virtual lines to create a dual flexible spider model. One or more new corner points are created on fulfillment of pre-defined conditions. Angle and magnitude of flow vectors is determined using one or more new corner points and alerts are given to the user based on the angle and magnitude.

Abstract: Systems and methods for down-scaling are provided. In one example, a method for processing image data includes determining a plurality of output pixel locations using a position value stored by a position register, using the current position value to select a center input pixel from the image data and selecting an index value, selecting a set of input pixels adjacent to the center input pixel, selecting a set of filtering coefficients from a filter coefficient lookup table using the index value, filtering the set of source input pixels to apply a respective one of the set of filtering coefficients to each of the set of source input pixels to determine an output value for the current output pixel at the current position value, and correcting chromatic aberrations in the set of source input pixels.

Abstract: A method and apparatus for measuring a distance to an object, be a device (100) having at least two cameras (104, 106), is described. One or more first images including the object are acquired by a first camera of the device and one or more first reference images are acquired by a second camera of the device, while the device is in a first position. One or more second images including the object and one or more second reference images are acquired by cameras of the device, while the device is in a second position, different from the first position. Based on the first and second reference images, information on the displacement of at least one camera of the device between the first and second position are determined. The distance from the device to the object is calculated based on the first and second images including the object and the determined information on the displacement of the at least one camera.

Abstract: Systems and methods for de-identification of medical images can be applied to medical images acquired using various techniques. A 3D medical image can be analyzed to generate an image mask that partitions the image into a foreground region and a background region. From the image mask, a “skin surface” can be reconstructed based on the boundary between the foreground region and the background region. The image mask can be modified, e.g., by moving a randomly-selected subset of the voxels from the foreground region to the background region so that the shape of the skin surface is altered, thus obscuring patient-identifying features. The original medical image can be modified by changing the intensity of voxels in the background region while preserving the original intensity of voxels in the foreground region.

Abstract: Techniques for tagging virtualized content are disclosed. In some embodiments, a modeled three-dimensional scene of objects representing abstracted source content is generated and analyzed to determine a contextual characteristic of the scene that is based on a plurality of objects comprising the scene. The modeled scene is tagged with a tag specifying the determined contextual characteristic.

Abstract: An observation apparatus includes an imaging unit, a driving mechanism, and a control section. The imaging unit includes an imaging section and an illumination section. The imaging section includes an image sensor and an imaging optical system, and images a sample to output an image signal. The illumination section includes a plurality of emitting sections which are located away from an optical axis of the imaging optical system and configured to emit illumination light toward the sample. The driving mechanism moves the imaging unit. The control section controls operations of the imaging section, the illumination section, and the driving mechanism. The control section determines a lighting emitting section of the emitting sections and causes the lighting emitting section to light based on the image signal, when the imaging unit is moved by the driving mechanism.

Abstract: Machine vision methods and systems determine if an object within a work field has one or more predetermined features. Methods comprise capturing image data of the work field, applying a filter to the image data, in which the filter comprises an aspect corresponding to a specific feature, and based at least in part on the applying, determining if the object has the specific feature. Systems comprise a camera system configured to capture image data of the work field, and a controller communicatively coupled to the camera system and comprising non-transitory computer readable media having computer-readable instructions that, when executed, cause the controller to apply a filter to the image data, and based at least in part on applying the filter, determine if the object has the specific feature. Robotic installation methods and systems that utilize machine vision methods and systems also are disclosed.

Abstract: An apparatus for determining a concentration of a driver includes: a region-of-interest definition processor configured to define a region-of-interest based on traffic-related information on front image information of a vehicle running; a driver eye-gaze detection processor configured to detect a plurality of eye-gaze points from driver image information of a driver in accordance with an eye-gaze movement of the driver; a driver concentration information calculation processor configured to calculate concentration information of each of the eye-gaze points; an image coordinate system processor configured to convert the region-of-interest to an image coordinate system and convert positions of the eye-gaze points to the image coordinate system; and a concentration determination processor configured to determine a concentration of the driver on the image coordinate system based on the concentration information.