Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for performing facial recognition. In one aspect, a method includes accessing a first digital photograph. A first face template is generated for each face detected in the first digital photograph. Second user identifiers that are associated with a first user identifier are determined. A digital photograph index of photographs, user identifiers, and areas in the digital photographs in which a face of a user identified by user identifier is located is accessed. Second user identifiers are selected, and second face templates are generated from the faces of the user the digital photographs. First face templates that match second face templates are identified, and for each first face template that matches a second face template, data is generated specifying the area in the first digital photograph in which the face of the second user is located.

Abstract: An object of the present invention is to provide a pattern-measuring apparatus and a semiconductor-measuring system which are able to obtain an evaluation result for suitably selecting processing with respect to a semiconductor device. In the present invention for attaining the object described above, there is proposed a pattern-measuring apparatus including an arithmetic device which compares a circuit pattern of an electronic device with a reference pattern, in which the arithmetic device classifies the circuit pattern in processing unit of the circuit pattern on the basis of a comparison of a measurement result between the circuit pattern and the reference pattern with at least two threshold values.

Abstract: An identity document digital image correction method comprising identifying the type, class and issue of a presented identity document from features in a digital image thereof; retrieving from a reference identity document repository, information about a reference authentic version of an identity document of the same type, class and issue as the presented identity document; allowing a user to select at least two features in the digital image of the presented identity document calculating the distance between the two selected features; determining the distance between corresponding features in the reference authentic version of the identity document; calculating a scaling factor ?x, y from the ratio of the distance between the two selected features in the presented identity document and the distance between corresponding features in the reference authentic version of the identity document; applying the scaling factor to measurements of features in the digital image of the presented identity document.

Abstract: A system and method that determines multiple linear groups of horizontally placed items in an image is disclosed. The method includes receiving an image including a portion of a shelving unit, identifying a plurality of items in the image, determining a plurality of facings, clustering the plurality of facings, merging a plurality of clusters and determining a list of linear groups in the image.

Abstract: The invention relates to a method for recording individual three-dimensional optical images (1) to form a global image (50) of a tooth situation comprising an upper jaw (2) and a lower jaw (3). A first 3D model (6) of a first subsection of the upper jaw and a second 3D model (8) of a second subsection of the lower jaw are produced from the individual images. Subsequently, a geometric positional relationship between the first 3D model (6) and the second 3D model (8) is determined, said positional relationship being determined by using a lateral image (20, 21, 22) and/or using a contact pattern (31, 32). Said lateral image (20, 21, 22) comprises an image area which comprises at least part of the first subsection of the upper jaw (2) and at least part of the second subsection of the lower jaw (3). Said contact pattern comprises several contact areas (31, 32) between the upper jaw and the lower jaw. Said contact pattern (30, 31, 32) is measured by means of an occlusion paper (29).

Abstract: The present invention provides apparatuses and systems for determining block averages in film grain simulation including determining block averages during a display process and determining block averages during a decoding process. The apparatuses and systems of the present invention exhibit different characteristics in terms of memory requirements and computational cost. More specifically, the first apparatus and system use no external memory, but require either extra reads of the blocks, or internal memory in the display pipeline, while the second apparatus and system require extra memory bandwidth and extra external memory (e.g., RAM).

Abstract: The present disclosure relates to a method and a system for reducing functional ambiguity from an image. In one embodiment, an input image is received and processed to identify objects. Spatial proximity score of the identified objects are determined based on which functional proximity score of functionalities associated with the identified objects is further determined. Upon determining the functional proximity score, possible domain of all the functionalities associated with the identified objects is determined. Further, a domain score is determined based on which the ambiguity of the domain related to the input image is reduced. A text summary of objects, functionalities and possible domains associated with the input image is then generated upon mapping with one or more user profiles and displayed to end user.

Abstract: An authentication device may include a housing and a finger sensor carried by the housing and including first processing circuitry and a finger sensing area coupled thereto. The first processing circuitry may be configured to generate finger image data based upon a finger positioned adjacent the finger sensing area, and generate and store a first template based upon the finger image data. The authentication device may include second processing circuitry carried by the housing and configured to obtain the finger image data from the first processing circuitry. The second processing circuitry may be configured to generate a second template based upon the finger image data. The first processing circuitry may further be configured to obtain the second template from second processing circuitry, and validate the second template against the first template.

Abstract: One embodiment of the invention provides an image processing method for use in fitting a deformable shape model to an acquired image. The deformable shape model specifies a set of target points whose motion is governed by the model. The method comprises, for each target point, generating a corresponding response image by: providing a feature detector to locate a given target point within the acquired image, wherein said feature detector includes a random forest comprising one or more decision trees; scanning a patch image across the acquired image to define multiple sampling locations with respect to the acquired image; for each sampling location of the patch image with respect to the acquired image, performing regression voting using the random forest to produce one or more votes for the location of the given target point within the acquired image; and accumulating the regression votes for all of the multiple sampling locations to generate said response image corresponding to the given target point.

Abstract: Systems and methods of generating a compact visual vocabulary are provided. Descriptor sets related to digital representations of objects are obtained, clustered and partitioned into cells of a descriptor space, and a representative descriptor and index are associated with each cell. Generated visual vocabularies could be stored in client-side devices and used to obtain content information related to objects of interest that are captured.

Abstract: A computer-implemented method that includes: accessing data encoding a fingerprint slap scan image, the fingerprint slap scan image including fingerprints of a number of fingers from a subject's hand; performing a radon transform of the fingerprint slap scan image to generate a sinogram, the sinogram represented on a grid having a lateral span axis and a projection angle axis; automatically identifying patterns of segments on the sinogram as finger lines, each finger lines corresponding to a particular finger present on the fingerprint slap scan image; and based on identified finger lines on the sinogram, automatically determining whether one or more fingers are missing on the fingerprint slap scan image.

Abstract: A disparity deriving apparatus for deriving disparity of an object includes a calculator to calculate costs between a first reference area in a reference image and each one of corresponding areas corresponding to the first reference area in the comparison image, and costs between a second reference area, surrounding the first reference area in the reference image, and each one of corresponding areas corresponding to the second reference area in a comparison image; a synthesizer to synthesize the costs of the first reference area and the second reference area as synthesis costs; a deriving unit to derive a disparity value of the object captured in the first reference area based on the synthesized synthesis costs; and a correction unit to apply a spatial filter to a local region, composed of a plurality of areas, in the first reference image to correct a disparity value of the local region.

Abstract: A fingerprint recognition device and method thereof are provided. The fingerprint recognition method includes: obtaining a plurality of swiping frames; generating a plurality of registered fingerprint datasets according to the swiping frames; obtaining a pressing fingerprint dataset; and comparing the pressing fingerprint dataset with the registered fingerprint datasets.

Abstract: A novel disparity computation technique is presented which comprises multiple orthogonal disparity maps, generated from approximately orthogonal decomposition feature spaces, collaboratively generating a composite disparity map. Using an approximately orthogonal feature set extracted from such feature spaces produces an approximately orthogonal set of disparity maps that can be composited together to produce a final disparity map. Various methods for dimensioning scenes and are presented. One approach extracts the top and bottom vertices of a cuboid, along with the set of lines, whose intersections define such points. It then defines a unique box from these two intersections as well as the associated lines. Orthographic projection is then attempted, to recenter the box perspective. This is followed by the extraction of the three-dimensional information that is associated with the box, and finally, the dimensions of the box are computed. The same concepts can apply to hallways, rooms, and any other object.

Abstract: A method and an apparatus for dividing an image area are provided. The method includes: a reading step: reading a pixel value of a current pixel and pixel values of multiple adjacent pixels adjacent to the current pixel in an image; a calculation step: calculating a difference between the pixel values of the current pixel and each adjacent pixel, so as to obtain a pixel difference value set, the pixel difference value set including multiple pixel difference values; a processing step: comparing each pixel difference value with a corresponding dynamic threshold in sequence, and determining all adjacent pixels in a same sub-image area as the current pixel, the current pixel and all the adjacent pixels of the current pixel in the same sub-image area constituting one sub-image area; and a traversal step: traversing each pixel in the image, and performing the reading step, the calculation step, and the processing step on each pixel in sequence, so as to acquire multiple sub-image areas of the image.

Abstract: In one embodiment, a method determines known features for existing face tracks that have identity labels and builds a database using these features. The face tracks may have multiple different views of a face. Multiple features from the multiple faces may be taken to build the face models. For an unlabeled face track without identity information, the method determines its sampled features and finds labeled nearest neighbor features with respect to multiple feature spaces from the face models. For each face in the unlabeled face track, the method decomposes the face as a linear combination of its neighbors from the known features from the face models. Then, the method determines weights for the known features to weight the coefficients of the known features. Particular embodiments use a non-linear weighting function to learn the weights that provides more accurate labels.

Abstract: An embodiment includes an ultrasonic sensor system comprising: a backend material stack including a first metal layer between a substrate and a second metal layer with each of the first and second metal layers including a dielectric material; a ultrasonic sensor including a chamber, having a negative air pressure, that is sealed by first and second electrodes coupled to each other with first and second sidewalls; an interconnect, not included in the sensor, in the second metal layer; wherein (a) a first vertical axis intersects the substrate, the chamber, and the first and second electrodes, (b) a second vertical axis intersects the interconnect and the substrate, (c) a first horizontal axis intersects the chamber, the interconnect, and the first and second sidewalls, and (d) the first and second electrodes and the first and second sidewalls each include copper and each are included in the second metal layer.

Abstract: Various systems and methods for tracking and analyzing emotional engagement of people while shopping are described herein. A system for tracking and analyzing emotional engagement of people while shopping includes an image processing module to: access image data from an electronic data storage, the image data including a person in a retail environment; and detect the person in the image data; an emotion detection module to determine an emotion expressed by the person; an object detection module to determine an object with which the person is interacting; and a marketing module to: select a sales action based on the emotion and the object; and execute the sales action.

Abstract: The present invention relates to the technical field of image processing, and in particular, to a patch net model for image representation and a construction method of the patch net model. The patch net model for image representation is of a forest-shaped structure consisting of a plurality of composite nodes and basic nodes, each composite node is a non-leaf node, and each basic node is a leaf node; the basic node includes a certain patch region of an image and a representative patch representing an apparent feature of the patch region; the composite node includes a certain patch region of the image and can be further decomposed into basic nodes and/or composite nodes; an edge exists between two nodes, which are located on the same layer of the forest-shaped structure and are spatially connected, and a relation matrix used for expressing the spatial relative position of the two nodes is arranged on the edge.

Abstract: An image coding method for dividing an input image into a plurality of divisional blocks having different sizes, and coding the image for each of the divisional blocks includes acquiring attribute information of a target block, setting a control parameter for controlling an image quality of the target block, determining a prediction control parameter based on the attribute information, calculating a difference value between the control parameter and the prediction control parameter, and coding the calculated difference value.