Abstract: A device is trained for face recognition. A first acquired digital image of a scene includes a face. Face image data is extracted and stored in a face image library along with an unique identifier. A second acquired digital image may or may not include the face of the same person as the face in the first acquired digital image. Face recognition is applied to extracted face data of the second digital image, and the face of the first digital image is displayed as a match to the face of the second digital image when the first and second images are determined to match or the second digital image or a face therein is identified with biometric data stored along with the first digital image when the faces in the first and second images are determined to match.

Abstract: A method for providing improved foreground/background separation in a digital image of a scene is disclosed. The method comprises providing a first map comprising one or more regions provisionally defined as one of foreground or background within the digital image; and providing a subject profile corresponding to a region of interest of the digital image. The provisionally defined regions are compared with the subject profile to determine if any of the regions intersect with the profile region. The definition of one or more of the regions in the map is changed based on the comparison.

Abstract: Different distances of two or more objects in a scene being captured in a video conference are determined by determining a sharpest of two or more color channels and calculating distances based on the determining of the sharpest of the two or more color channels. At least one of the objects is identified as a foreground object or a background object, or one or more of each, based on the determining of the different distances. The technique involves blurring or otherwise rendering unclear at least one background object or one or more portions of the scene other than the at least one foreground object, or combinations thereof, also based on the determining of distances.

Abstract: An adaptive motion estimation and deblurring technique for acquired digital images includes acquiring multiple digital images with a moving digital image acquisition device that includes an image sensor, including a relatively sharp, underexposed reference image and a blurred image. Anb initial approximate point spread function (PSF) is estimated corresponding to the moving of the device. A different DC offset point is determined and a second PSF is calculated based on the different DC offset point.

Abstract: Sub-regions within a face image are identified to be enhanced by applying a localized smoothing kernel to luminance data corresponding to the sub-regions of the face image. An enhanced face image is generated including an enhanced version of the face that includes certain original pixels in combination with pixels corresponding to the one or more enhanced sub-regions of the face.

Abstract: A forward interpolation approach is disclosed for enabling a second version of an image to be constructed from a first version of the image. According to one implementation, pixels from the first version of the image are mapped to pixels in the second version of the image, and pixel values are determined for the corresponding pixels in the second version of the image. In one implementation, the pixel mapping is performed using a lookup table and linear approximation. Performing the pixel mapping in this manner enables computations to be simplified and cost and gate count to be reduced.

Abstract: A method and device is provided for adjusting the white balance of a digital image by adjusting the values assigned to the red, green, and blue subpixels of a pixel in the image. The adjustment to the subpixels is determined by identifying pixels in the image that have an RGB product greater than a threshold value, wherein the threshold value is based at least in part on an average of the RGB products of each pixel in the image and a variance between the RGB products of the pixels and the average of the RGB products.

Abstract: Methods and systems for detecting and correcting chromatic aberration and purple fringing are disclosed. Chromatic aberration can be addressed by separating an image into color planes and then adjusting these to reduce chromatic aberration by using a specific calibration image (calibration chart) as an empirical method to calibrate the image acquisition device. Purple fringing can be corrected by initially addressing color aberration resulting from the lateral chromatic aberration (LCA). The LCA is first removed and then the correction is extended to purple fringing. A discovery is relied upon that the purple fringing is created in the direction of the chromatic aberration and is more pronounced in the direction of the chromatic aberration.

Abstract: A method and system for image recognition in a collection of digital images includes training image classifiers and retrieving a sub-set of images from the collection. For each image in the collection, any regions within the image that correspond to a face are identified. For each face region and any associated peripheral region, feature vectors are determined for each of the image classifiers. The feature vectors are stored in association with data relating to the associated face region. At least one reference region including a face to be recognized is/are selected from an image. At least one classifier on which said retrieval is to be based is/are selected from the image classifiers. A respective feature vector for each selected classifier is determined for the reference region. The sub-set of images is retrieved from within the image collection in accordance with the distance between the feature vectors determined for the reference region and the feature vectors for face regions of the image collection.

Abstract: A portable digital image acquisition device includes multiple lenses and/or multiple flashes. A main digital image and first and second reference images are acquired. The first and second reference images are acquired with different flash-lens combinations that have different flash-lens distances. One or more flash eye defects are detected and corrected in the main image based on analysis of the first and second reference images.

Abstract: A digital image processing technique gathers visual meta data using a reference image. A main image and one or more reference images are captured on a hand-held or otherwise portable or spatial or temporal performance-based image capture device. The reference images are analyzed based on predefined criteria in comparison to the main image. Based on said analyzing, supplemental meta data are created and added to the main image at a digital data storage location.

Abstract: A method of processing a digital image using face detection within the image achieves one or more desired image processing parameters. A group of pixels is identified that correspond to an image of a face within the digital image. Default values are determined of one or more parameters of at least some portion of the digital image. Values are adjusted of the one or more parameters within the digitally-detected image based upon an analysis of the digital image including the image of the face and the default values.

Abstract: A method of adjustable spatial and/or temporal compression of an image including a face includes identifying a group of pixels that correspond to a face within a digitally-acquired image. A first compression portion of the image including the group of pixels is determined. A second compression portion of the image other than the group of pixels is also determined. The first compression portion may be automatically compressed with higher-grade compression than the second compression portion to generate a compressed image including the face, or an option to provided the compressed image including the different grade compressions may be provided.

Abstract: A method for red-eye detection in an acquired digital image acquiring one or more preview or other reference images without a flash. Any red regions that exist within the one or more reference images are determined. A main image is acquired with a flash of approximately a same scene as the one or more reference images. The main image is analyzed to determine any candidate red eye defect regions that exist within the main image. Any red regions determined to exist within the one or more reference images are compared with any candidate red eye defect regions determined to exist within the main image. Any candidate red eye defect regions within the main image corresponding to red regions determined also to exist within the one or more reference images are removed as candidate red eye defect regions.

Abstract: A microelectronic unit is provided in which front and rear surfaces of a semiconductor element may define a thin region which has a first thickness and a thicker region having a thickness at least about twice the first thickness. A semiconductor device may be present at the front surface, with a plurality of first conductive contacts at the front surface connected to the device. A plurality of conductive vias may extend from the rear surface through the thin region of the semiconductor element to the first conductive contacts. A plurality of second conductive contacts can be exposed at an exterior of the semiconductor element. A plurality of conductive traces may connect the second conductive contacts to the conductive vias.

Abstract: A method and apparatus for adjusting the effect of applying a kernel to a signal is provided. The adjustment may be based on the magnitude of an effect that a kernel matrix is intended to have on the signal matrix. The adjustments can be based on factors including, but not limited to, a signal-to-noise ratio of the signal data, properties of a device (e.g., lens) used to capture the signal data, or a metric that is derived based on an analysis of the signal data. Processing image data in accordance with an embodiment of the present invention is aimed at regaining lost contrast that may be due to properties of optics that are used to capture the image data.

Abstract: A technique of generating a panoramic image involves acquiring a set of at least two main image frames, as well as corresponding generating and/or acquiring corresponding relatively low-resolution image frames, of overlapping portions of a panoramic scene. A map or other information is stored relating to the generation of a low-res panorama. A main panorama image is formed by joining main image frames based on the map or other information gained in the low-res process.

Abstract: A method for detecting a redeye defect in a digital image containing an eye comprises converting the digital image into an intensity image, and segmenting the intensity image into segments each having a local intensity maximum. Separately, the original digital image is thresholded to identify regions of relatively high intensity and a size falling within a predetermined range. Of these, a region is selected having substantially the highest average intensity, and those segments from the segmentation of the intensity image whose maxima are located in the selected region are identified.

Abstract: A digital image acquisition system includes a portable apparatus for capturing digital images and a digital processing component for detecting, analyzing and informing the photographer regarding motion blur, and for reducing camera motion blur in an image captured by the apparatus. The digital processing component operates by comparing the image with at least one other image, for example a preview image, of nominally the same scene taken outside the exposure period of the main image. In one embodiment the digital processing component determines the degree of artifacts and whether to inform the user that the image is blurred by identifying at least one feature in a single preview image which is relatively less blurred than the corresponding feature in the main image.

Abstract: An unsatisfactory scene is disqualified as an image acquisition control for a camera. An image is acquired. One or more eye regions are determined. The eye regions are analyzed to determine whether they are blinking, and if so, then the scene is disqualified as a candidate for a processed, permanent image while the eye is completing the blinking.