Abstract: An image processing apparatus and a method for improving the coding efficiency for a quantization parameter are provided. The method includes steps of setting a predicted quantization parameter for a current coding unit by using multiple quantization parameters which are set for multiple surrounding coding units located around the current coding unit which is target of coding processing, and setting a difference quantization parameter indicating a difference value between the quantization parameter which is set for the current coding unit and the predicted quantization parameter.

Abstract: Briefly, embodiments of methods and/or systems of detecting and image of a human face in a digital image are disclosed. For one embodiment, as an example, parameters of a neural network may be developed to generate object labels for digital images. The developed parameters may be refined by a neural network to generate signal sample value levels corresponding to probability that a human face may be depicted at a localized region of a digital image.

Abstract: The invention provides a method for instantly recognizing and positioning an object, comprising steps of: a) wirelessly searching a wireless identification of the object; b) capturing a plurality of images of the object for each image capture; c) determining a 2D center coordinate (x,y) of the object based on a center coordinate (xw,yw) of the wireless identification of the object; d) transforming the captured images of the object to acquire a 3D pattern of the object, and comparing the 3D pattern of the object with 3D patterns pre-stored; and e) if the 3D pattern of the object matches with a pre-stored 3D pattern, calculating and obtaining a 3D center coordinate (x,y,z) of the object to recognize and position the object. The invention also provides a system and a processor enabling the method, and use of the system.

Abstract: A head mounted device (HMD) includes an imaging unit configured to capture at least one image of a partial region of an iris, an electrocardiogram (ECG) sensor configured to receive an ECG signal, and a control unit configured to acquire at least one image of the partial region of the iris and ECG signals, and authenticate a user by using the acquired image(s) of the partial region of the iris and the ECG signals.

Abstract: Embodiments disclose methods, systems and non-transitory computer readable medium for joint temporal segmentation and classification of user activities in an egocentric video.

Abstract: An optical module includes a housing with a lighting mechanism, an aperture formed therein, and a window that frames a transparent surface adapted to contact a skin surface of a person, for example, a fingertip. The lighting mechanism provides light to illuminate the skin surface placed upon the transparent surface. A prism has a first side facing the lighting mechanism, a second side at the window, and a third side through which a dactylographic image exits the prism. One or more light reflecting surfaces are disposed within the housing to reflect the dactylographic image towards the housing aperture. The optical module is coupled to a mobile device having a camera, with the aperture of the housing aligning with a lens of the camera. The camera acquires the dactylographic image, and the mobile device adjusts this dactylographic image to produce a dactylographic image suitable for biometric matching.

Abstract: Systems, methods, and articles of manufacture for expense management are disclosed. A user may submit receipts for one or more transactions into an expense management system. The expense management system may extract information from the receipts and match the extracted information to one or more of a plurality of Records of Charge (ROCs). The expense management system may match the extracted information to a ROC based on vendor name, location, time stamp, and the like. The expense management system may further assign a ROC to an expense category. The expense management system may assign a ROC to an expense category based on a standard industrial classification (SIC) code of a merchant, a category assigned to a different ROC, and a location associated with the ROC. An expense report for all submitted receipts may be generated.

Abstract: An apparatus for detecting a feature in an image includes an image input section for receiving at least part of the image in the form of image data having a plurality of pixels, the plurality of pixels comprising a plurality of non-border pixels, a feature detection module adapted to attribute a feature probability value to each of the pixels of the image data, and an extremum determination module for determining at least one local extremum among the feature probability values, wherein the extremum determination module is adapted to output, for each of the plurality of pixels, a final indication if the feature probability value of the pixel in question is a local extremum.

Abstract: This disclosure is about a method and apparatus for estimating heading misalignment between a device and a person, where the device/apparatus comprises an optical sensor, the optical sensor capable of capturing an image of the person. At least one physical feature of the person is extracted from the image. A misalignment angle is estimated using the at least one physical feature of the person extracted from the image.

Abstract: According to an embodiment, an image processing apparatus includes an interface that receives a first image and a second image. A processor extracts a first character string appearing on the first image, and a second character string appearing on the second image. The processor determines that both the first character string and the second character string correspond to a photographed subject in the first image and second image. The processor cuts out each of the characters constituting the first character string, in units of a character. The processor cuts out each of the characters constituting the second character string, in units of a character. The processor calculates a first similarity degree group and a second similarity degree group. The processor recognizes a character string of the subject, based on the first similarity degree group and the second similarity degree group.

Abstract: An itinerary estimation device includes a classification unit and an estimation unit. The classification unit classifies multiple pieces of expense data including information related to expenses into one or more expense data groups, according to a chronological order of date information extracted from each of the multiple pieces of expense data. The estimation unit estimates an itinerary that includes at least dates on the basis of information including date information extracted from each piece of classified expense data included in the expense data group.

Abstract: A computer-implemented system and method for spine group generation and arrangement are provided. A set of spine groups each having one or more spines of clusters and at least one singleton cluster is obtained. A set of concentric rings are designated in a display and at least a portion of the spine groups are placed within the concentric rings. An innermost ring includes one or more of the spine groups that are unique from the other spine groups. A primary ring includes a majority of the spine groups placed within the display and an outermost ring delineates placement of descriptive labels for the placed spine groups.

Abstract: An apparatus and method of estimating an image optical flow are provided. The method includes receiving temporally-consecutive first and second images; calculating rates of change of brightness values of pixels in the received temporally-consecutive first and second images; calculating a first optical flow estimation result value of a first pixel within the first image and a second optical flow estimation result value of a second pixel within the second image by using the calculated rates of change; comparing the first optical flow estimation result value with the second optical flow estimation result value; and correcting the first optical flow estimation result value and the second optical flow estimation result value by using a result of the comparison.

Abstract: Briefly, embodiments of methods and/or systems of training multiclass convolutional neural networks (CNNs) are disclosed. For one embodiment, as an example, an auxiliary CNN may be utilized to form an ensemble with the collection as a linear combination. The linear combination may be based, at least in part, on boost prediction error encountered during the training process.

Abstract: The present invention discloses a method for identifying PDF document. wherein, it comprises the following steps: S1: analyzing the path objects in the PDF document, and identifying the forms in PDF document; S2: analyzing the text objects outside the form regions in the PDF document, and recognizing the text contents in the PDF document; S3: writing the identified results into a temporary file, or writing them into the PDF document as an attachment. The method for identifying PDF document provided by the present invention could identify the tables, the paragraphs, titles, the tabulations and so on in the PDF document, thereby, the PDF document can be edited with the paragraph as a unit, and be tagged conveniently to confirm the reading order, so as to facilitate the reading of people with visual impairment; in the same time, it also can derive document in other forms according to the identified results, which thereby greatly facilitates users to read and edit the PDF document.

Abstract: Image processing apparatus includes a local amount generation unit, a correspondence point calculation unit, a relative correspondence point information calculation unit, a correspondence point selecting unit and a decision unit. The local feature value generation unit calculates, for a first image, a set of information about first local feature values including a first feature point(s). The correspondence point calculation unit calculates, for a second image, as information about correspondence point(s), a correspondence relation between the first and second feature point(s) contained in a set of information about second local feature values calculated from the second image.

Abstract: An image processing apparatus and a method for improving the coding efficiency for a quantization parameter are provided. The method includes steps of setting a predicted quantization parameter for a current coding unit by using multiple quantization parameters which are set for multiple surrounding coding units located around the current coding unit which is target of coding processing, and setting a difference quantization parameter indicating a difference value between the quantization parameter which is set for the current coding unit and the predicted quantization parameter.

Abstract: A handwriting recognition system converts word images on documents, such as document images of historical records, into computer searchable text. Word images (snippets) on the document are located, and have multiple word features identified. For each word image, a word feature vector is created representing multiple word features. Based on the similarity of word features (e.g., the distance between feature vectors), similar words are grouped together in clusters, and a centroid that has features most representative of words in the cluster is selected. A digitized text word is selected for each cluster based on review of a centroid in the cluster, and is assigned to all words in that cluster and is used as computer searchable text for those word images where they appear in documents. An analyst may review clusters to permit refinement of the parameters used for grouping words in clusters, including the adjustment of weights and other factors used for determining the distance between feature vectors.

Abstract: A smart refrigerator has a body having a shelf and a drawer, a door coupled to the body and has a plurality of trays. The shelf has a corresponding camera configured to capture an image of items on the shelf. The drawer has a corresponding camera configured to capture an image of items in the drawer. The trays have a corresponding camera configured to capture an image of items in the trays. The images captured by the cameras are processed to identify the type of the item and the location of the item within the smart refrigerator.

Abstract: A method of reducing image resolution in a deep convolutional network (DCN) includes dynamically selecting a reduction factor to be applied to an input image. The reduction factor can be selected at each layer of the DCN. The method also includes adjusting the DCN based on the reduction factor selected for each layer.