Abstract: A voice control method and display apparatus are provided. The voice control method includes converting a voice of a user into text in response to the voice being input during a voice input mode; performing a control operation corresponding to the text; determining whether speech of the user has finished based on a result of the performing the control operation; awaiting input of a subsequent voice of the user during a predetermined standby time in response to determining that the speech of the user has not finished; and releasing the voice input mode in response to determining that the speech of the user has finished.

Abstract: Methods, devices and apparatus for reconstructing an image are provided. In an aspect, a method includes: obtaining coincidence events detected by crystal pairs in a detector of a PET device by scanning a phantom, reconstructing a phantom image according to the coincidence events, and obtaining a time calibration amount of each of crystals in the detector by determining a crystal based source position deviation and a crystal based source time difference of a crystal pair including the crystal according to the phantom image, determining a time difference of the crystal pair according to respective time differences for the crystal pair detecting a plurality of coincidence events, pre-calibrating the time difference of the crystal pair with the crystal based source time difference, and obtaining the time calibration amount of the crystal according to the pre-calibrated time difference of the crystal pair.

Abstract: Techniques are described for controlling display of video data and/or other image data based at least in part on selective mapping of pixel values to pixels. Such techniques may include separating a display panel into multiple regions, with at least one primary region having a highest resolution of displayed image data and with one or more secondary regions having one or more lower resolutions of displayed image data (e.g., by using a 1-to-M mapping of image pixel values to display panel pixels for each such secondary region, where M is greater than 1, such that each such image pixel value controls the display of M such pixels). The image data may further be encoded and optionally decoded in accordance with such a display panel arrangement, such as to encode the image data per the display panel arrangement to reduce its size before transmission to the display panel.

Abstract: An image processing apparatus and a control method thereof are provided.

Abstract: Provided are an image synthesis apparatus, an image synthesis method, and a program thereof, capable of determining a combination of a synthesis candidate image and a background image of which impressions match each other. Plural synthesis candidate images are input, and impression values of the plural synthesis candidate images are determined with respect to plural impression axes. Further, plural background images are input, and impression values of the plural background images are determined with respect to the plural impression axes. A combination of a synthesis candidate image and a background image having a small difference between impression values is determined. The synthesis candidate image and the background image of the determined combination are synthesized to generate a synthetic image.

Abstract: An abnormality detection apparatus includes a feature extraction unit configured to extract an image feature according to a common algorithm, a flow calculation unit, a first abnormality detection unit, and a second abnormality detection unit. An extraction range for the image feature is composed of a predetermined first partial area in a first image, a predetermined second partial area in a second image, and areas near places in the first and second images predicted as destinations of the feature point. The first abnormality detection unit detects an abnormality in the first (second) image based on an optical flow for a feature point in the first (second) partial area. The second abnormality detection unit detects an abnormality by using a feature point in a first overlapped extraction area defined in a first overlapped area and a feature point in a second overlapped extraction area defined in a second overlapped area.

Abstract: Predicting patch displacement maps using a neural network is described. Initially, a digital image on which an image editing operation is to be performed is provided as input to a patch matcher having an offset prediction neural network. From this image and based on the image editing operation for which this network is trained, the offset prediction neural network generates an offset prediction formed as a displacement map, which has offset vectors that represent a displacement of pixels of the digital image to different locations for performing the image editing operation. Pixel values of the digital image are copied to the image pixels affected by the operation by: determining the vectors pixels that correspond to the image pixels affected by the image editing operation and mapping the pixel values of the image pixels represented by the determined offset vectors to the affected pixels.

Abstract: A system, method and computer-readable storage devices for providing unsupervised normalization of noisy text using distributed representation of words. The system receives, from a social media forum, a word having a non-canonical spelling in a first language. The system determines a context of the word in the social media forum, identifies the word in a vector space model, and selects an “n-best” vector paths in the vector space model, where the n-best vector paths are neighbors to the vector space path based on the context and the non-canonical spelling. The system can then select, based on a similarity cost, a best path from the n-best vector paths and identify a word associated with the best path as the canonical version.

Abstract: Disclosed is a co-segmentation method and apparatus for a three-dimensional model set, which includes: obtaining a super patch set for the three-dimensional model set which includes at least two three-dimensional models, each of the three-dimensional models including at least two super patches; obtaining a consistent affinity propagation model according to a first predefined condition and a conventional affinity propagation model, the consistent affinity propagation model being constraint by the first predefined condition which is position information for at least two super patches that are in the super patch set and belong to a common three-dimensional model set; converting the consistent affinity propagation model into a consistent convergence affinity propagation model; clustering the super patch set through the consistent convergence affinity propagation model to generate a co-segmentation outcome for the three-dimensional model set.

Abstract: A method and device for detecting objects from scene images by using dynamic knowledge base is disclosed. The method includes segmenting a scene image captured by at least one camera into a plurality of image segments. Each of the plurality of image segments include a plurality of pixels. The method further includes detecting at least one object in each of the plurality of image segments. The method includes finding a plurality of similar images from an online database based on the at least one object. The method further includes identifying using a knowledge base engine, at least one similar object in the plurality of similar images. The method includes updating the knowledge base engine based on the at least one similar object identified from the plurality of similar images. The method further includes training a neural network to detect objects from scene images, based on the updated knowledge base engine.

Abstract: A character recognition device includes an acquisition means configured to acquire an image containing a character region, a first recognition means configured to recognize a character from the character region by a first recognition method, a setting means configured to set reference lines along an alignment direction of the characters and passing through a specified position in each character, a second recognition means configured to recognize a character by a second recognition method, the second recognition method being a method that recognizes a character from an image within a recognition window by scanning in a recognition target region in an image while changing a size of the recognition window, and configured to set a position or a height in a vertical direction of the recognition window based on the reference lines, and an output means configured to output a word composed of characters recognized by the second recognition means.

Abstract: A method of training a learning machine to detect spills of hydrocarbon liquids from pipelines. A neural network is trained by collecting samples of a number of different ground materials as well as a number of liquid hydrocarbons. For each hydrocarbon, a spill is simulated on each ground material. For each of these spills, a thermal camera and a visible light camera are used to capture images. The images from the two cameras are fused, and input to the neural network for classification training. Once the neural network is trained, a system having the two cameras and the neural network can be used to detect actual hydrocarbon spills.

Abstract: An apparatus includes a first brightness value acquisition unit configured to obtain a first absolute brightness value by converting a representative brightness value of an image obtained using an image pickup unit, into absolute brightness, a first determination unit configured to determine an output value based on an input-output characteristic of an output device, the output value corresponding to the first absolute brightness value obtained by the first brightness value acquisition unit, and a second determination unit configured to determine information about exposure corresponding to the output value.

Abstract: System that facilitates rapid onboarding of an autonomous (cashier-less) store by capturing images of items in the store's catalog from different angles, with varying backgrounds and lighting conditions, and that automatically builds a classifier training dataset from these images. The system may have cameras in different positions, lights supporting variable illumination, and monitor screens that generate different background colors. It may have an input device such as a barcode reader, and an operator terminal that prompts operators to place items into the imaging system in the necessary orientations. Once an item is placed in the imaging system, a fully automated process may generate a sequence of background colors, a sequence of lighting conditions, and may capture and process images from all of the cameras to create training images. Training images for an item may be generated in seconds, compared to many minutes per item using manual image capture and processing.

Abstract: A watermark image may be generated that includes a first set of encoded pixels each of which is assigned a first transparency value and a second set of encoded pixels each of which is assigned a second transparency value, the second transparency level being different from the first transparency level. The encoded pixels may be distributed among a set of blank pixels such that each encoded pixel neighbors one or more blank pixels in the watermark image, and in particular at least two blank pixels in the watermark image. Herein, each blank pixel may be assigned the second transparency value. The watermark image may be overlaid and blended over a background source image to create an encoded source image. A decoder system may recover encoded information from the encoded source image.

Abstract: Digital image completion using deep learning is described. Initially, a digital image having at least one hole is received. This holey digital image is provided as input to an image completer formed with a framework that combines generative and discriminative neural networks based on learning architecture of the generative adversarial networks. From the holey digital image, the generative neural network generates a filled digital image having hole-filling content in place of holes. The discriminative neural networks detect whether the filled digital image and the hole-filling digital content correspond to or include computer-generated content or are photo-realistic. The generating and detecting are iteratively continued until the discriminative neural networks fail to detect computer-generated content for the filled digital image and hole-filling content or until detection surpasses a threshold difficulty.

Abstract: An image compression device includes: a first compression processing circuit configured to generate a first compressed image data by performing a first block compression process on an image data and a dither value generator circuit. When the first compressed image data associated with a compression target block is generated, the dither value generator circuit is configured to generate at least one dither value in response to block coordinates indicating a position of the compression target block in an image and the first compression processing circuit is configured to perform a quantization process using the at least one dither value in the first block compression process.

Abstract: A mechanism is provided for improving predicate parses (or logical representations of a passage) using semantic knowledge. In response to encountering an ambiguous decision point during a syntactic analysis of a portion of natural language content, a candidate meaning of the ambiguous decision point is generated. Characteristics of the ambiguous decision point are evaluated based on a semantic knowledge base to determine a semantic meaning associated with the ambiguous decision point. A determination is made as to whether the semantic meaning supports or refutes the candidate meaning. In response to determining that the semantic meaning refutes the candidate meaning, the candidate meaning of the ambiguous decision point is overridden based on the semantic meaning to include the semantic meaning as a final meaning for the ambiguous decision point. The portion of natural language content is then processed based on the final meaning for the ambiguous decision point.

Abstract: An image acquisition apparatus, an image forming apparatus, and a control method of the same are provided. An image acquisition apparatus includes an image sensor in which a red element outputting a signal corresponding to a red light, a green element outputting a signal corresponding to a green light, and a blue element outputting a signal corresponding to a blue light are arranged in a predetermined pattern, a driving unit configured to move the image sensor to a plurality of positions; and an image processor configured to receive a signal output from the red element, the green element and the blue element in the plurality of positions, respectively, and configured to acquire an image of an object by combining the received signal.

Abstract: Systems and methods for reconstructing a document from a series of document images. An example method comprises: receiving a plurality of image frames, wherein each image frame of the plurality of image frames contains at least a part of an image of an original document; identifying a plurality of visual features in the plurality of image frames; performing spatial alignment of the plurality of image frames based on matching the identified visual features; splitting each of the plurality of image frames into a plurality of image fragments; identifying one or more text-depicting image fragments among the plurality of image fragments; associating each identified text-depicting image fragment with an image frame in which that image fragment has an optimal value of a pre-defined quality metric among values of the quality metric for that image fragment in the plurality of image frames; and producing a reconstructed image frame by blending image fragments from the associated image frames.