Abstract: A search device identifies names of POI from a document group having not been determined. A storage unit that stores a POI presence/absence learning model having learned contexts relating to presence/absence of POI, a POI state learning model having learned contexts relating to states of POI, and a POI name learning model having learned features relating to names of POI, an acceptance unit that accepts a first document group that is a determination target, first and second determination units and an identifying unit that identifies a name of a POI using the POI name learning model from each document of a third document group for which information relating to states of POI is determined by the second determination unit in a second document group are included.

Abstract: Mechanisms for identification of text fields in documents using neural networks are described. Identification of text fields includes obtaining a plurality of symbol sequences of a document having a plurality of text fields, determining a plurality of vectors representative of one of the plurality of symbol sequences, processing the plurality of vectors using a first neural network to obtain, based on values of the plurality of vectors, a plurality of recalculated vectors, determining an association between a first recalculated vector of the plurality of recalculated vectors and a first text field of the plurality of text fields, the first recalculated vector being representative of a first symbol sequence of the plurality of symbol sequences, and determining, based on the association between the first recalculated vector and the first text field, an association between the first symbol sequence and the first text field.

Abstract: Systems and methods for processing video are provided. The method includes receiving a text-based description of active scenes and representing the text-based description as a word embedding matrix. The method includes using a text encoder implemented by neural network to output frame level textual representation and video level representation of the word embedding matrix. The method also includes generating, by a shared generator, frame by frame video based on the frame level textual representation, the video level representation and noise vectors. A frame level and a video level convolutional filter of a video discriminator are generated to classify frames and video of the frame by frame video as true or false. The method also includes training a conditional video generator that includes the text encoder, the video discriminator, and the shared generator in a generative adversarial network to convergence.

Abstract: Embodiments disclosed herein may comprise receiving a run-time image of a run-time die and, with a deep learning module, identifying a characteristic noise in the run-time image, and modifying the run-time image to reduce the characteristic noise, thereby generating a de-noised run-time image. Such embodiments may be performed as methods, by systems, or from non-transitory computer-readable storage media on one or more computing devices. An image sensor of a metrology tool may capture the run-time image of the run-time die. The metrology tool may include a run-time die disposed on a specimen, a run-time image sensor, and a processor in electronic communication with the run-time image sensor. Embodiments may further comprise receiving a training image of a training die, modifying the training image, and training the deep learning module to identify the characteristic noise in the run-time image and modify the run-time image.

Abstract: Techniques are described for content-adaptive downsampling of digital images and videos for computer vision operations, such as semantic segmentation. A computer vision system comprises a memory, one or more processors operably coupled to the memory and a downsampling module configured for execution by the one or more processors to perform, based on a non-uniform sampling model trained to predict content-aware sampling parameters, downsampling input image data to generate downsampled image data. A segmentation module is configured for execution by the one or more processors to perform segmentation on the downsampled image to produce a segmentation result, such as a feature map that assigns pixels of the downsampled image data to object classes. An upsampling module is configured for execution by the one or more processors to perform upsampling according to the segmentation result to produce upsampled image data.

Abstract: An example apparatus including memory to store a first image of a document and a second image of the document, and a processor coupled to the memory, wherein the processor is to: perform optical character recognition on the first image to generate a first output dataset; perform optical character recognition on the second image to generate a second output dataset; and compute a transformation matrix based on the first output dataset and the second output dataset, the transformation matrix to align the first image with the second image.

Abstract: Disclosed is a depth map calculation method and apparatus. The depth map calculation method includes calculating a global sparse depth map corresponding to a current frame using frames including the current frame, calculating a local dense depth map corresponding to the current frame using the current frame, extracting a non-static object region from the current frame by masking a static object region, removing the non-static object region from the global sparse depth map, and generating a global dense depth map corresponding to the current frame by merging the non-static object region-removed global sparse depth map and the local dense depth map.

Abstract: A method for object tracking includes: obtaining frames with a number of N of history images of the object; acquiring first predicted feature point information of each frame image by using first network models corresponding to each frame image in the frames with a number of N of history images, and acquiring second predicted feature point information of the each frame image by using second network models corresponding to each frame image; adjusting parameters of the first network model and parameters of the second network model based on the first predicted feature point information and the second predicted feature point information until the first network model and the second network model are trained completely; and performing tracking of the object by using the first completely trained network model and the second completely trained network model.

Abstract: Embodiments allow live action images from an image capture device to be composited with computer generated images in real-time or near real-time. The two types of images (live action and computer generated) are composited accurately by using a depth map. In an embodiment, the depth map includes a “depth value” for each pixel in the live action image. In an embodiment, steps of one or more of feature extraction, matching, filtering or refinement can be implemented, at least in part, with an artificial intelligence (AI) computing approach using a deep neural network with training. A combination of computer-generated (“synthetic”) and live-action (“recorded”) training data is created and used to train the network so that it can improve the accuracy or usefulness of a depth map so that compositing can be improved.

Abstract: A method and a system for generating training images for training an instance segmentation machine learning algorithm (MLA). A set of image-level labelled images are received, where a given image is labelled with a label indicative of a presence of an object having an object class in the image. A classification MLA detects the object having the object class in each image. A class activation map (CAM) indicative of discriminative regions used by the classification MLA for detecting the object in each image is generated. A region proposal MLA is used to generate region proposals for each image. A pseudo mask of the respective object is generated based on the region proposals and the CAM, where a pseudo mask is indicative of pixels corresponding to the respective object class. The pseudo masks are used as a label with the image-level labelled images for training the instance segmentation MLA.

Abstract: There is provided a system comprising a processor configured to obtain a set of images of a semiconductor specimen, (1) for an image of the set of images, select at least one algorithmic module MS out of a plurality of algorithmic modules, (2) feed the image to MS to obtain data DMS representative of one or more defects in the image, (3) obtain a supervised feedback regarding rightness of data DMS, (4) repeat (1) to (3) for a next image until a completion criterion is met, wherein an algorithmic module selected at (1) is different for at least two different images of the set of images, generate, based on the supervised feedback, a score for each of a plurality of the algorithmic modules, and use scores to identify one or more algorithmic modules Mbest as the most adapted for providing data representative of one or more defects in the set of images.

Abstract: Methods and systems for generating synthetic point cloud data are described. Projected 2D data grid is generated by projecting a 3D point cloud into a 2D grid, with rotation equivariance. A generative model is learned using the projected 2D data grid, wherein the generative model is implemented using flex-convolution and transpose flex convolution operations, for example in a generative adversarial network. The learned generative model is used to generate synthetic point clouds.

Abstract: A pattern inspection apparatus includes an optical image acquisition mechanism to acquire optical image data of a plurality of regions from a substrate where a plurality of figure patterns are formed, a plurality of comparison circuits each of which performs one of die-to-die inspection processing for comparing the optical image data with each other and die-to-database inspection processing for comparing the optical image data with reference image data generated from design pattern data, and an inspection circuit to individually output, for each region of the plurality of regions, the optical image data of a region concerned to comparison circuits, whose number is variably set for each region, in the plurality of comparison circuits, and to control each comparison circuit, serving as an output destination of the optical image data in the plurality of comparison circuits, to perform one of the die-to-die inspection processing and the die-to-database inspection processing.

Abstract: Example systems and methods of selection of video frames using a machine learning (ML) predictor program are disclosed. The ML predictor program may generate predicted cropping boundaries for any given input image. Training raw images associated with respective sets of training master images indicative of cropping characteristics for the training raw image may be input to the ML predictor, and the ML predictor program trained to predict cropping boundaries for raw image based on expected cropping boundaries associated training master images. At runtime, the trained ML predictor program may be applied to a sequence of video image frames to determine for each respective video image frame a respective score corresponding to a highest statistical confidence associated with one or more subsets of cropping boundaries predicted for the respective video image frame. Information indicative of the respective video image frame having the highest score may be stored or recorded.

Abstract: An estimation method implemented by a computer, the estimation method includes: executing learning processing by training an autoencoder with a data group corresponding to a specific task; calculating a degree of compression of each part regarding data included in the data group by using the trained autoencoder; and estimating a common part with another piece of data included in the data group regarding the data corresponding to the specific task based on the calculated degree of compression of each part.

Abstract: Voltage supply system with boost converter and charge pump. A voltage supply system can include a boost converter controllable to receive an input voltage at an input node and generate an output voltage when the output voltage is greater than or equal to the input voltage. The voltage supply system can include a charge pump controllable to receive the input voltage at the input node and generate the output voltage when the output voltage is less than the input voltage. The voltage supply system can further include a controller configured to receive a control signal and control the boost converter or the charge pump to generate the output voltage at an output node based on the control signal.

Abstract: A method for classifying depth scan data at a computing device includes: obtaining, at the computing device, a set of depth measurements and a graphical representation of the depth measurements; automatically selecting, at the computing device, a subset of the depth measurements indicating a region of interest; rendering, on a display of the computing device, an image including (i) the graphical representation of the depth measurements and (ii) a graphical indication of the region of interest overlaid on the graphical representation of the depth measurements; receiving, via an input device, a selection associated with the image; and generating a region of interest indicator based on the subset of the depth measurements and the selection.

Abstract: Techniques of adjusting the salience of an image include generating values of photographic development parameters for a foreground and background of an image to adjust the salience of the image in the foreground. These parameters are global in nature over the image rather than local. Moreover, the optimization of the salience over such sets of global parameters is provided through two sets of these parameters by an encoder: one set corresponding to the foreground, in which the salience is to be either increased or decreased, and the other set corresponding to the background. Once the set of development parameters corresponding to the foreground region and the set of development parameters corresponding to the background region have been determined, a decoder generates an adjusted image with an increased salience based on these sets of development parameters.

Abstract: Devices, systems, and methods of wirelessly controlling appliances and electronic devices, such as ceiling fans, air conditioners, garage doors, or the like. A receive-only garage door system is wirelessly controlled by a proprietary remote control unit. A cloning unit is able to clone or duplicate the proprietary wireless signal, and to replay it or re-generate it in response to a triggering command that a user submitted via a smartphone or tablet; thereby enabling to control such garage door system via mobile electronic devices. The cloning unit utilizes recording of the wireless signal payload and carrier frequency; wireless signal analysis; image analysis of the appliance or of the remote control unit; queries to a remote server to obtain properties of the proprietary wireless signal; or other techniques of signal analysis or duplication.

Abstract: The invention provides a wafer map identification method, which includes the following steps: obtaining a wafer map of at least one to-be-identified wafer; performing an image processing operation on the wafer map and a reference pattern, wherein the image processing operation includes: performing a convolution operation on the wafer map and the reference pattern respectively, extracting a critical feature of the wafer map after the convolution operation, and calculating a weight distribution based on the reference pattern after the convolution operation; and calculating a similarity between the processed wafer map and the processed reference pattern to identify the wafer map. The invention also provides a computer-readable recording medium recording the above identification method.