Abstract: A computing system comprising a generative neural network (100) is disclosed. The generative neural network (100) is configured to receive a plurality of input videos. Each input video comprises a face defining an identity. Each input video comprises a behaviour. Each input video comprises the same behaviour as each of the other input videos, and a different identity to each of the other input videos. The generative neural network (100) is also configured synthesize an output video from the input videos. The output video comprises a synthetic face defining a synthetic identity. The generative neural network has been trained, with a loss function, to preserve the behaviour of the input videos while generating the synthetic identity that is different from each identity of the input videos.

Abstract: A method for providing map data which include position information for first landmarks of a first landmark class, collecting environment data, and determining a position of the mobile unit. Training data are generated and stored for the first landmarks based on the position of the mobile unit, the collected environment data and the position information. Based on the training data, a first detector module is generated for detecting the first landmark class. The position determination system includes a memory unit for providing map data which include position information for first landmarks of a first landmark class, a data acquisition unit for collecting environment data, a localization unit for determining a position of the mobile unit, and a processing unit for generating and storing training data based on the position of the mobile unit, the collected environment data and the position information for the first landmarks.

Abstract: A separation unit that generates a separated image in which a plurality of structures are separated, from an image including the plurality of structures receives an input of an image pair that includes a target image relating to at least a part of the plurality of structures and a non-separation image not including the structure, to output a separation image in which one of the structures is extracted from the target image. The separation unit receives an input of a new image pair including the target image and the separation image, to output a new separation image in which another one of the structures is extracted from the target image. The separation unit repeats the reception of the input of the new image pair including the target image and the new separation image and the output of a new separation image in which another one of the structures is extracted from the target image.

Abstract: Disclosed is a computer-implemented method of segmenting a medical patient image using an atlas and relating the segmentation result to a model of possible geometric changes to the segmentation result (e.g. for correcting the position of the segmentation of anatomical structures) which consider for example anatomical limitations. The thus-related segmentation result may be used as a basis for changing and/or correcting the position, shape and/or orientation of at least parts of the segmentation result, e.g. by user interaction. The invention also relates to an atlas data set comprising information such as values of the variables of the model of possible geometric changes in relation to the positions of anatomical structures in the atlas.

Abstract: An optical condition determination system includes a simulation execution unit that performs an optical simulation on a surface texture model that models a surface texture of a target object of the appearance inspection, and a defect model that models a defect of the target object, under a plurality of optical conditions to generate a surface texture image and a defect image, an image synthesizing unit that synthesizes the surface texture image and the defect image generated by an optical simulation under the same optical condition to generate a synthetic image, an evaluation value calculating unit that calculates an evaluation value representing easiness of detecting the defect in the synthetic image, a correlation analysis unit that analyzes a correlation between an optical condition and the evaluation value corresponding to the synthetic image, and an optimum condition searching unit that searches for the optical condition suitable for the appearance inspection based on an analysis result of the correlatio

Abstract: A focusing control device includes: a sensor that has a pair of a first signal detection unit that receives one of a pair of beams passed through different parts disposed in one direction in a pupil region of an imaging optical system including a focus lens and detects a signal depending on an intensity of received light and a second signal detection unit that receives other of the pair of beams and detects a signal depending on an intensity of received light, in which a plurality of the pairs that are arranged in the one direction form a pair line and a plurality of the pair lines are arranged in a direction orthogonal to the one direction; an addition correlation operation unit as defined herein; a non-addition correlation operation unit as defined herein; a selection unit as defined herein; and a lens driving section as defined herein.

Abstract: For scanning optical patterns, such as two-dimensional QR codes, with a mobile device at increased distances, a first image is acquired. A region of interest likely containing the optical pattern in the first image is identified. The mobile device then zooms in on the region of interest and a second image is acquired. The optical pattern is then decoded using the second image.

Abstract: The present invention relates to a method for generating a biometric signature of a subject comprising: obtaining a plurality of sequential video frame images of a moving subject from a video segment; obtaining a portion of each frame comprising a surrounding of the moving subject; carrying out a transformation function to the frequency domain on one or more of said portions of the frames comprising a surrounding of a of the subject; and optionally saving the spectral characteristics of said transformation function in a repository. The present invention also relates to a system for carrying out said method.

Abstract: The present invention relates to a method for viewing a specimen using a microscope which comprises an objective lens and an image sensor for converting an image formed on the image sensor by the objective lens. A field of view of the microscope can be varied by selecting a section of the image sensor. In one step of the method, an initial image of at least a partial section of the specimen is captured with the microscope, for which a first field of view is selected on the microscope. The initial image is analyzed to determine at least two differing fields of view forming a partial area image, wherein a partial area of the initial image is formed by each of the fields of view forming a partial area image. Images of the partial areas of the specimen are captured for each of the determined fields of view forming a partial area image. The invention further relates to a microscope for viewing a specimen using a microscope.

Abstract: A medical image processing device includes circuitry configured to: calculate relative information including relative positional relation of a pathologically diagnosed position in a subject with respect to an observation image of the subject captured by an imager, by using data on the observation image and pathological diagnosis information on the subject; and generate, based on the relative information, data on a displayed observation image resulting from addition of the pathological diagnosis information to the data on the observation image.

Abstract: According to one embodiment, a gradient estimation device includes one or more processors. The processors acquire a first image and a second image obtained by photographing a surface of an object. The processors extract, from the first image, a first pixel indicating a boundary of an object in a direction different from a gradient direction of the surface, and extract a second pixel corresponding to the first pixel from the second image. The processors calculate gradient magnitude of the surface by transforming at least one of the first image and the second image such that a matching degree between the first pixel and the second pixel is increased.

Abstract: The embodiment of the disclosure provides an unlocking control method and related products. The method includes: detecting whether the electronic device is in a motion state; reducing a first recognition threshold to obtain a second recognition threshold when the electronic device is in a motion state; acquiring first biometric information when a matching value between first biometric information and first preset biometric template information is greater than the second recognition threshold; and acquiring second biometric information and performing an recognition operation on the second biometric information. The method can improve the efficiency of biometric recognition when the electronic device is in motion.

Abstract: A camera system for use in a mobile handheld computing and communication device with a front side with a display screen and a back side has two separate and independent cameras positioned on the back side of the device, wherein the cameras have a physical separation between them of substantially the same separation as between the eyes on a human face and, wherein the camera system creates a 3D-Like image of an object for storage and for viewing on a display screen.

Abstract: An information processing apparatus includes an acquisition unit that acquires process information associated with a reference image and regarding an input process performed by referencing the reference image, and a process decision unit that decides a content of a next input process to perform by referencing the reference image, on a basis of the process information acquired by the acquisition unit.

Abstract: A pose-invariant visual speech recognition system obtains a single view input of a speaker, such as a single video stream captured by a single camera. The single view input provides a particular pose of the speaker, which refers to a view angle, relative to the lens or image capture component of the camera that captured the video of the speaker, at which the speaker's face is captured. The pose of the speaker is used to select a visual speech recognition model to use to generate a text label that is the words spoken by the speaker. One or more additional view angles of the speaker are also generated from the single view input of the speaker. These one or more additional view angles, along with the single view input of the speaker, are used by the selected visual speech recognition model to generate the text label for the speaker.

Abstract: A method and device for geo-referencing of supplied aerial image data, which has been recorded by an imaging system and in which a part of the earth's surface is imaged, wherein the geo-referencing of the aerial image data is carried out based on the supplied geo-referenced synthetic aperture radar (SAR) image data, in which the same part of the earth's surface is imaged, wherein the method and system include: determination of essentially point-like radar signatures RS in the geo-referenced SAR image data, provision of a position of the sun for recording time of the aerial image data and the part of the earth's surface imaged therein, determination of elongated cast shadows SS produced in a direction of solar radiation in the aerial image data based on the position of the sun at the time of recording, determination in the aerial image data of initial points AP facing the sun of the determined cast shadows SS, determination of spatially corresponding point-like radar signatures RS and the initial points AP of

Abstract: Real-time autofocus. In an embodiment, a scanning apparatus includes an imaging sensor, a focusing sensor, an objective lens, and processor(s) configured to analyze image data captured by the imaging and focusing sensors, and move the objective lens. Real-time autofocus during scanning of a sample is achieved by determining a true-Z value for the objective lens for a point on a sample and for each of a plurality of regions on the sample. The true-Z values and/or surfaces calculated therefrom are used to determine a predicted-Z value for an unscanned region of the sample. The objective lens is adjusted to the predicted-Z value at the beginning of the unscanned region. After scanning the region, a true-Z value is determined for the region and compared to the predicted-Z value. A rescan of the region is initiated if the comparison exceeds a predetermined threshold.

Abstract: For patient positioning for scanning, a current pose of a patient is compared to a desired pose. The desired pose may be based on a protocol or a pose of the same patient in a previous examination. Any differences in pose, such as arm position, leg position, head orientation, and/or torso orientation (e.g., laying on side, back, or stomach), are communicated. By changing the current pose of the patient to be more similar to the desired pose, a more consistent and/or registerable dataset may be acquired by scanning the patient.

Abstract: An apparatus and method are described for capturing images in visible light as well as other radiation wavelengths. In one embodiment, the apparatus comprises: a diffraction coded imaging system including a plurality of apertures arranged in a diffraction coded array pattern with opaque material blocking array elements not containing apertures; and a light- or radiation-sensitive sensor coupled to the diffraction coded imaging system array and positioned at a specified distance behind the diffraction coded imaging system array, the radiation-sensitive sensor configured to sense light or radiation transmitted and diffracted through the apertures in the diffraction coded imaging system array.

Abstract: The disclosure relates to devices, systems and methods for generating a digital image of a tissue section that is a composite of two or more source digital images of adjacent tissue sections and which may have a real-time adjustable boundary between different source images. The devices include computer software products for a fused-view visualization tool which permits one or more of generating and displaying the composite image and modifying the location of one or more boundaries between source images comprising the composite image. The systems include computer-implemented systems such as work stations and networked computers for analyzing tissue samples using the fused-view visualization tool. The methods include processes for visualization of a tissue sample as a composite image derived from two or more slides of adjacent tissue sections, for example as an interactive composite image wherein the proportion of each source image in the composite image may be altered.

Abstract: A control apparatus (105) includes a normalizer (400, 401) that performs normalization processing on a first signal and a second signal by using normalization coefficients related to the first signal and the second signal, respectively, a correlation calculator (402) that performs correlation calculation with respect to the normalized first and second signals, and a corrector (403, 800, 801) that corrects correlation data to cancel the normalization processing, and the correlation data is based on an output signal from the correlation calculator.

Abstract: At least one focus detection signal processing unit calculates a first correlation amount obtained by comparing a pair of image signals with each other in a same direction on a time axis and a second correlation amount obtained by comparing the pair of image signals with each other in opposite directions on the time axis, the pair of image signals being output from an image sensor and respectively corresponding to a pair of light beams that have each passed through a corresponding one of areas, in an exit pupil of an imaging optical system, that are at least partially different from each other. A camera control unit determines whether an image shift amount corresponding to the first correlation amount corresponds to an in-focus position, based on the first correlation amount and the second correlation amount.

Abstract: Certain embodiments of the present disclosure relate to a computer-implemented method and system which include patch-wise coregistration (PWCR) for change detection using remote sensing images which can be taken from satellites, aircraft, UAV and other platforms, where the images can be nadir or off-nadir images and can be acquired from the same or different view-angles. The remote sensing images can be bi-temporal or multi-temporal. VHF satellite images can be used.

Abstract: A method for processing multiple image sources comprises steps: receiving a plurality of first images from a first device; arranging the first images in an order of times of capturing the first images; receiving at least one second image from at least one second device; arranging together the second image and the most spatially neighboring first images according to first and second positioning information of the first and second images; arranging the first and second images sharing an identical geographic space into an identical geographic group; arranging the first and second images, which have an identical still background, into an identical background group; sorting the first and second images of each background group in an order from a large background to a small background or in a reverse order; and processing the sorted images into a video. The present invention makes viewers watching the video feel smooth without abruptness.

Abstract: A method for the simultaneous imaging of different physical properties of an examined medium from the simultaneous joint inversion of multiple datasets of physical field measurements is described. The method introduces Gramian spaces of model parameters and/or their transforms, and Gramian constraints computed as the determinants of the corresponding Gram matrices of the model parameters and/or their transforms. Gramian constraints are introduced as additional regularization terms, and their minimization enforces the correlation between different model parameters and/or their transforms. The method does not require a priori knowledge about specific analytical or empirical or statistical correlations between the different model parameters and/or their attributes, nor does the method require a priori knowledge about specific geometric correlations between different model parameters and/or their attributes.

Abstract: Computer techniques and systems for decomposing a scene into small scene partitions that can be independently solved and merged into a single cohesive scene without loss in reconstruction quality.

Abstract: Automatically scanning multiple document sheets with a camera includes receiving a video stream while the camera is pointed at the multiple document sheets, detecting presence of a first top page of the multiple document sheets based on the video stream, taking a still photograph of the first top page in response to detecting presence of the first top page, detecting presence of a second top page based on the video stream by confirming that the second top page is different from the first top page and by waiting a predetermined amount of time for an image of the second top page to stabilize, and taking a still photograph of the second top page in response to detecting presence of the second top page. Detecting the pages may include determining that the camera is pointing at the stack of documents and a detected page is not obstructed.

Abstract: An image alignment method and apparatus, where the method and apparatus include obtaining image information of two to-be-aligned images, determining, using a cross-correlation measurement model, first coordinate offset according to the image information of the two images, where the first coordinate offset are used to indicate position deviations of to-be-aligned pixels between the two images in the coordinate system, and aligning the two images according to coordinates of pixels in the first image in the coordinate system and the first coordinate offset.

Abstract: Systems and techniques are provided for measuring consistency of animation and/or video rendering on a device, and providing a user with an indication of one or more consistency scores of the user's device. The consistency score indicates generally the consistency of user experience that is expected for the user device. In some cases, the consistency score is calculated based upon the differences in rendering times between adjacent frames in an animation, and/or the number of frames that have a render difference that differs from a standard or baseline value. A consistency score also may be based upon the framerate achievable by the device when rendering a video at a constant quality.

Abstract: An imaging apparatus includes an imaging element, a readout unit, and a signal processing unit. The imaging element includes pixels that have photoelectric conversion units for one micro lens. The readout unit performs a first readout operation on pixels included in a first region of the imaging element to read signals according to accumulated electric charges, and performs a second readout operation on pixels included in a second region different from the first region to read signals according to accumulated electric charges. The signal processing unit makes a correction for reducing noise levels to the signals readout by the readout unit. The signal processing unit makes the correction to the signals acquired from the first region such that differences between noise levels included in the signals read by the first readout operation and the second readout operation become smaller after the correction by the signal processing unit.

Abstract: A biological signal detection device including a first FFT transformer that performs fast Fourier transformation on a signal of a first sensor, and that outputs a complex data row formed by a first real part data row and a first imaginary part data row; a second FFT transformer that performs fast Fourier transformation on a signal of a second sensor, and that outputs a complex data row formed by a second real part data row and a second imaginary part data row; a frequency component subtractor that subtracts the first real part data row from the second real part data row to output a noise-removed first real part data row; and a FFT inverse transformer that performs fast Fourier inverse transformation on the noise-removed first real part data row and the second imaginary part data row.

Abstract: Disclosed are systems and methods for improving interactions with and between computers in content searching, generating, hosting and/or providing systems supported by or configured with personal computing devices, servers and/or platforms. The systems interact to identify and retrieve data within or across platforms, which can be used to improve the quality of data used in processing interactions between or among processors in such systems. The disclosed systems and methods provide systems and methods for automatically determining and suggesting media for a user based on a determined state of the user. The state determination of a user is based on user input respective to media representing specific categories of moods.

Abstract: A system and method that allows the capture of a series of images to create a single linear panoramic image is disclosed. The method includes capturing an image, dynamically comparing a previously captured image with a preview image on a display of a capture device until a predetermined overlap threshold is satisfied, generating a user interface to provide feedback on the display of the capture device to guide a movement of the capture device, and capturing the preview image with enough overlap with the previously captured image with little to no tilt for creating a linear panorama.

Abstract: A method, apparatus, and computer program product for generating a seamless and error-free panorama image using a selective set of views from a multi-view image from each of the several captured light field (LF) images. The method identifies the view and corresponding image for each captured LF image which has the same or closely located center of projection with the views of neighboring LF images. Image registration and warping techniques are applied across the images and the parallax error is calculated which indicates the closeness of their center of projections. The view from each LF captured image with minimal parallax error is selected and stitched together with the other views identified as having minimal parallax error.

Abstract: A particular implementation determines objective distortion levels (d,) respectively for a plurality of artifact types. The objective distortion levels are aligned to ensure that the same distortion level of different types of artifacts corresponds to the same perceived distortion level. The aligned distortion levels (d,?) are sorted to obtain sorted distortion levels (d,?). The sorted distortion levels are then pooled together into an overall distortion level or an overall quality metric. The sorted distortion levels may be pooled using a weighted sum, wherein the weight is larger when the sorted distortion level is greater.

Abstract: A system and method that allows the capture of a series of images to create a single linear panoramic image is disclosed. The method includes capturing an image, dynamically comparing a previously captured image with a preview image on a display of a capture device until a overlap threshold is satisfied, generating a user interface to provide feedback on the display of the capture device to guide a movement of the capture device, and capturing the preview image with enough overlap with the previously captured image with little to no tilt for creating a linear panorama.

Abstract: Limitations of conventional dimensional printing techniques are addressed to provide features and flexibility not presently available, including extracting images selectively from a Postscript® or PDF file and thus enable texturing of individual images within a page; constructing the texture automatically directly from the image using image processing techniques; visualizing the texture to be applied to an image via construction of a bump map or normal map in openGL and DirectX®; adjusting texturing parameters via visual feedback in openGL and DirectX®; and inserting a clear texture back into a PDF file for printing automatically.

Abstract: A system and method that allows the capture of a series of images to create a single linear panoramic image is disclosed. The method includes capturing an image, dynamically comparing a previously captured image with a preview image on a display of a capture device until a predetermined overlap threshold is satisfied, generating a user interface to provide feedback on the display of the capture device to guide a movement of the capture device, and capturing the preview image with enough overlap with the previously captured image with little to no tilt for creating a linear panorama.

Abstract: Embodiments generally relate to providing systems and methods for assessing image quality of a distorted image relative to a reference image. In one embodiment, the system comprises a convolutional neural network that accepts as an input the distorted image and the reference image, and provides as an output a metric of image quality. In another embodiment, the method comprises inputting the distorted image and the reference image to a convolutional neural network configured to process the distorted image and the reference image and provide as an output a metric of image quality.

Abstract: Methods and systems for detecting defects on a wafer are provided. One method includes altering one or more design clips based on how the one or more design clips will appear in output generated by a wafer inspection process for a wafer. The method also includes aligning the one or more altered design clips to the output generated for the wafer during the wafer inspection process. In addition, the method includes detecting defects on the wafer based on the output aligned to the one or more altered design clips.

Abstract: A method is provided, which comprises generating at least three images of an area of interest from at least one imaging system, the generated images being provided from at least three different angles, establishing point correspondence between the provided images. The method further involves generating at least two sets of three-dimensional information based on the provided images, wherein the at least two sets of three-dimensional information are generated based on at least two different combinations of at least two of the at least three provided images of the area of interest. The method further includes comparing the at least two sets of three-dimensional information so as to determine discrepancies, and providing information related to the imaging system or errors in the images based on the determined discrepancies.

Abstract: A system and method that allows the capture of a series of images to create a single linear panoramic image is disclosed. The method includes capturing an image, dynamically comparing a previously captured image with a preview image on a display of a capture device until a predetermined overlap threshold is satisfied, generating a user interface to provide feedback on the display of the capture device to guide a movement of the capture device, and capturing the preview image with enough overlap with the previously captured image with little to no tilt for creating a linear panorama.

Abstract: A system and method that allows the capture of a series of images to create a single linear panoramic image is disclosed. The method includes capturing an image, dynamically comparing a previously captured image with a preview image on a display of a capture device until a predetermined overlap threshold is satisfied, generating a user interface to provide feedback on the display of the capture device to guide a movement of the capture device, and capturing the preview image with enough overlap with the previously captured image with little to no tilt for creating a linear panorama.

Abstract: The interactive and shared surface technique described herein employs hardware that can project on any surface, capture color video of that surface, and get depth information of and above the surface while preventing visual feedback (also known as video feedback, video echo, or visual echo). The technique provides N-way sharing of a surface using video compositing. It also provides for automatic calibration of hardware components, including calibration of any projector, RGB camera, depth camera and any microphones employed by the technique. The technique provides object manipulation with physical, visual, audio, and hover gestures and interaction between digital objects displayed on the surface and physical objects placed on or above the surface. It can capture and scan the surface in a manner that captures or scans exactly what the user sees, which includes both local and remote objects, drawings, annotations, hands, and so forth.

Abstract: A system and method for tracking the position of a region in a subject using an ultrasound system is provided. Training data that indicates the position of the region is acquired from the subject while ultrasound data is acquired from the subject using the ultrasound system. A position mapping table is formed using the training data and ultrasound data; this table relates position information contained in the training data to the acquired ultrasound data. The position of the region is tracked by acquiring additional ultrasound data from the subject using the ultrasound system. This additional ultrasound data is compared to the ultrasound data contained in the position mapping table to identify the position of the region based on a unique relationship between the position of the region and ultrasound echo signatures of the ultrasound data acquired in both the training and motion tracking stages.

Abstract: A stereoscopic video temporal frame offset measurement system uses an improved signature curve generator to generate a robust signature curve for each of the left and right image sequences of a stereo video signal. The resulting signature curves are cross-correlated over a specified correlation range to generate cross-correlation coefficients. The cross-correlation coefficients are compared to a high threshold, and a temporal frame offset is generated and displayed for each cross-correlation coefficient that exceeds the high threshold. Each new temporal frame offset result is highlighted when displayed, otherwise the last result is displayed in a background fashion.

Abstract: The invention relates to a method of generating a synthesized image representing a view of a scene from a first input image representing the view and a second input image representing the view, the synthesized image comprising synthesized image positions, by assigning a synthesized image data value to a synthesized image position. The method comprises determining whether input images have at the synthesized image position data values associated with them. If either one has a data value associated, that data value is assigned to the synthesized image position. If both, an average of both values is assigned. The average is a weighed average, with weighing factors being a function of a distance or distances to a closest image position with no image data value or valid image data value assigned.

Abstract: A portable device for the storage and display of digital images, using an integrated projector for image display. The device itself might be a camera, smart phone or other portable or handheld electronic device capable of capture or storage of digital images. The integrated projector would allow projection of images at a larger size on an external surface than an internal display would permit. Certain embodiments of the device might also include image capture hardware, in addition to image storage, along with image editing or manipulation software. Method of display of digital images using a portable device with integrated projector is also disclosed.

Abstract: Target land area is three-dimensionally located using an optical sensor and a radar sensor in satellites. The positions of the satellites are measured and images are captured when the sensors point toward land areas. The aiming direction of the optical sensor for each area is measured. The distance between each area and the radar sensor pointing thereto is measured. Images of land areas captured by the sensors and covering common areas are matched. The distance between the target area selected from the common areas and the optical sensor is accurately determined on the basis of the positions of the sensors, the distance between the target area and the radar sensor, and the aiming direction of the optical sensor, which are associated with the images covering the target area.

Abstract: An apparatus for detecting a camera tampering, includes: an image capturing unit to capture at least one image; an input-edge-image generating unit to extract an edge image from an object displayed in the captured image and generate an input edge image by using the extracted edge image; a reference-edge-image generating unit to generate a reference edge image from the input edge image; a stolen-edge-image generating unit configured to generate a stolen edge image by substracting the input edge image from the reference edge image; and a tampering determining unit to compare the input edge image with the reference edge image, compare the reference edge image with the stolen edge image, and determine whether or not a camera tampering has occurred, based on a first similarity and a second similarity.