Abstract: The techniques described herein relate to methods, apparatus, and computer readable media configured to encode and/or decode video data. Point cloud video data is received that includes metadata specifying one or more regions of interest of the point cloud video data. A first region of interest is determined from the one or more regions of interest. A portion of the point cloud video data associated with the first region of interest is determined. Point cloud media is generated for viewing by a user based on the determined portion of the point cloud video data associated with the first region of interest.

Abstract: If wavelength defragmentation is performed during the operation of an optical network, an instantaneous interruption of a network arises; consequently, data are lost; therefore, an optical network control method according to an exemplary aspect of the present invention includes monitoring a data volume of a client signal to be transmitted using a plurality of optical subcarriers; and performing synchronously, depending on a variation in the data volume, an optical subcarrier changing process of changing an active optical subcarrier, of the plurality of optical subcarriers, to be used for transmitting the client signal, and a remapping process of remapping the client signal onto an active optical subcarrier after having been changed.

Abstract: A video super resolution method comprises successively executing instances of a first plurality of layers (SISR) of a neural network for generating a first image (St) at a higher resolution than an input image frame (Xt); successively executing a second plurality of layers (VSR) of the neural network for generating a second image (Vt) at the higher resolution, at least one of the second plurality of layers generating intermediate output information (Ht), the second plurality of layers taking into account an output image (Yt?1) at the higher resolution generated by a previous instance of the network from a previous input image frame (Xt?1) and intermediate output information (Ht?1) generated by the second plurality of layers of the previous instance, and executing a third plurality of layers for combining the first (St) and second (Vt) images to produce an output image (Yt) for the instance of the network.

Abstract: An optical network and a method are described. In the method, an orchestrator of an optical communication system receives an operation to execute, the operation being to activate or deactivate a service within a transmission signal of the optical communication system, the optical communication system having a span and an amplifier coupled to and supplying optical signals into each span. Network status data for each span within the optical communication system is retrieved, and the list of operations is analyzed with the network status data including existing data traffic on the fiber optic line to select a subset of the list of operations to execute that maintains the transmission signal below a bit error rate threshold. The orchestrator issues one or more signals to cause the one or more service within the subset of the list of operations to be activated or deactivated on the optical communication system.

Abstract: An electronic image capture device captures a first image of a scene at a first time. A first local tone mapping operator for a first portion of the first image is determined. The electronic image capture device further captures a second image of the scene at a second time. A motion of the electronic device between the first time and the second time is determined. A second local tone mapping operator for a second portion of the second image is determined. The second portion is determined to correspond to the first portion based, at least in part, on the determined motion of the electronic device. The second local tone mapping operator is determined based, at least in part, on the first local tone mapping operator. At least the second local tone mapping operator is applied to the second portion of the second image.

Abstract: A method and device for processing a light field video is described. The light field video includes a set of image views per unit of time, the light field video being associated with a scene without cuts. In the method a first super-rays representation of reference image views at a given time is determined based on centroids. A second super-rays representation associated with corresponding views of a subsequent set of image views is next determined based on de-projection and re-projection of centroids. The displacement of centroids between the first and second super-rays is determined and then the determined displacement is applied to centroids of the second super-rays representation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, are provided. In one aspect, a first image of an eye region of a subject is received, the first image captured under illumination by electromagnetic radiation having a wavelength in the 850 nm-1100 nm range. Based on the first image, a metric indicative of differential absorption of the electromagnetic radiation by an iris of the subject and a sclera of the subject is determined. It is identified, based on the metric, that the subject is a spoof representation of a human. Responsive to identifying that the subject is a spoof representation of a human, access to a secure system is prevented.

Abstract: Large digital images are classified by analyzing them at a subimage level and assigning classification probabilities to the subimages; these may be combined into a classification probability for the entire image. Classification probabilities may be visualized across the image using probabilities computed for the subimages. This enables ready identification of the image subregions upon which classification is based as well as the classification scores or probabilities associated therewith. For example, a large source image—too large to be analyzed directly by a neural network—may be decomposed into smaller subimages such as square tiles, which are sifted based on a visual criterion. The visual criterion may be image entropy, density, background percentage, or other discriminator. A neural network produces tile-level classifications that are aggregated to classify the source image, and overlapping tiles are used to create a probability map showing subimage probabilities.

Abstract: Systems and method are provided for detecting key-points in at least one image which involves dividing the at least one image into a plurality of image blocks; determining sparse representations of the image blocks using a dictionary D; determining at least one measure for each of the sparse representations of the image block; optionally determining key-points for which the at least one measure for each of the image blocks falls within a specified range; and retaining only the determined key-points with the at least one measure that is a local maxima in a sliding window when performing non-maxima suppression and identifying these retained key-points as detected key-points. In accordance with further teachings provided herein, there is also provided methods and systems for performing scale and/or rotation invariant sparse coding based key-point (SRI-SCK) detection and processing of images.

Abstract: An electronic device and method for image mask generation using a deep neural network is provided. The electronic device stores an input image that includes an object of interest generates a pilot mask for the input image by application of a trained deep neural network (DNN) on the input image. The electronic device extracts, from the input image, a region of interest that includes the object of interest, based on the generated pilot mask and resizes the extracted region of interest such that a size of the resized region of interest equals a threshold scaling size. The electronic device further generates a final mask by reapplication of the trained DNN on the resized region of interest.

Abstract: In a method of processing an image which is a video image including a plurality of frame images, a first high dynamic range (HDR) image is generated by combining a first frame image and a second frame image subsequent to the first frame image. The first and second frame images are included in the plurality of frame images. The first frame image has a first exposure time. The second frame image has a second exposure time different from the first exposure time. A second HDR image is generated by combining the second frame image and a third frame image subsequent to the second frame image. The third frame image is included in the plurality of frame images and has the first exposure time.

Abstract: The present invention provides a method for image preprocessing, the method including: identifying a plurality of input images to process image registration; selecting at least one transformation method among histogram equalization and gamma transformation in consideration of hue and brightness values of the plurality of input images; processing the histogram equalization on the plurality of input images in response to the histogram equalization method being selected; and processing the gamma transformation on the plurality of input images, in response to the gamma transformation method being selected.

Abstract: A controlled switch having N inputs and a single output (N?2) is switchable between N states. In each state a respective one of the inputs is connected to the single output. There are N sources of sub-streams of analog samples, each sub-stream composed of pairs of adjacent analog samples. Each source is coupled to a respective one of the inputs. In operation, the controlled switch is controlled by a control signal to switch between the N states. While the controlled switch is in any one of the states, a data transition occurs between two adjacent analog samples in the sub-stream whose source is coupled to the input that is connected to the single output. The single output yields a high-bandwidth analog signal. Any pair of adjacent analog samples in any one of the sub-streams substantially determines a corresponding pair of adjacent analog samples in the high-bandwidth analog signal.

Abstract: A computer-implemented method according to one embodiment includes identifying principal components for a dataset defined by data instances and features corresponding to the data instances, identifying, for at least one of the data instances, at least some of the principal components, wherein the identified principal components are determined to be salient for said at least one data instance, and determining, for said at least one of the data instances, one or more salient features corresponding to the identified salient principal components.

Abstract: A power over fiber system includes: a first data communication device including a power sourcing equipment device; a second data communication device including a powered device; and an optical fiber cable. The first data communication device and the second data communication device perform optical communication with one another. The first data communication device is capable of controlling low power supply and high power supply that are performed by the power sourcing equipment device. Feed electric power by the high power supply exceeds feed electric power by the low power supply. The first data communication device starts up the second data communication device by the low power supply to the second data communication device, and enables the high power supply after receiving predetermined light from the second data communication device, and disables the high power supply before receiving the predetermined light.

Abstract: Methods, systems and computer program products for providing improved visualization of inspection results are provided. Aspects include receiving an image from a connected camera by a computing device and processing the image by a graphics processing unit to divide the image into a plurality of image blocks. Responsive to inputting the processed image into a trained model, a plurality of image areas within the processed image is identified. Each image area is associated with an importance level and includes a unique set of image blocks. A sequence of image blocks is determined based on an importance level associated with each of the plurality of image blocks, wherein the sequence includes a list of image blocks in descending order by the importance level of each image block. The image blocks are stored in order of the sequence of image blocks by the server.

Abstract: A method and a device for determining a device parameter of a hearing device to be worn at the ear of a user, the method comprising: obtaining image data of one or more images, the image data including first image data of a first image of the outer human ear at a first angle; identifying, based on the first image data, a first anatomical landmark and a second anatomical landmark of the outer human ear using a processor; identifying the position of the first anatomical landmark and the second anatomical landmark in the first image; determining the device parameter of the hearing device based on the first anatomical landmark and the second anatomical landmark; and outputting the device parameter.

Abstract: A method of encoding at least one image divided into blocks, implementing: encoding a first syntax element associated with a characteristic of the at least one image; encoding, for a current block to be encoded associated with at least one encoding parameter, the data of the current block; if the first syntax element is encoded according to a first predefined value representing the characteristic of the image, a second syntax element representing the value of the at least one encoding parameter is encoded only if the current block belongs to a predefined zone of the image; if the first syntax element is encoded according to a second predefined value representing the characteristic of the image, the second syntax element is encoded, whether or not the current block belongs to the predefined zone.

Abstract: An example optical receiving device includes a photodiode to receive an optical signal, where the photodiode is configured to conduct a current that is based on an optical power of the optical signal, and a radio frequency (RF) gain circuitry to generate one or more analog electrical signals based on the current and based on gain provided by the RF gain circuitry. A power detector is configured to receive an analog electrical signal of the one or more analog electrical signals, to detect alternating current (AC) power of the optical signal based on the analog electrical signal, and to output a signal representing the AC power based on the detecting.

Abstract: A system for defect review and classification is disclosed. The system may include a controller, wherein the controller may be configured to receive one or more training images of a specimen. The one or more training images including a plurality of training defects. The controller may be further configured to apply a plurality of difference filters to the one or more training images, and receive a signal indicative of a classification of a difference filter effectiveness metric for at least a portion of the plurality of difference filters. The controller may be further configured to generate a deep learning network classifier based on the received classification and the attributes of the plurality of training defects. The controller may be further configured to extract convolution layer filters of the deep learning network classifier, and generate one or more difference filter recipes based on the extracted convolution layer filters.