Abstract: The horizontal and vertical offset for the M×N DCT block boundaries are determined for a decompressed image originally produced by a DCT-based compression system, regardless of the amount of compression and/or cropping that had been applied to the image. The method includes the steps of: a) computing one or more selected DCT coefficients from nonoverlapping, contiguous M×N blocks of the decompressed image, beginning at a selected offset; b) computing a coefficient histogram from a set of values for each selected DCT coefficient obtained from the blocks of the decompressed image; c) computing a dispersion metric from each coefficient histogram; d) repeating steps (a) through (c) for a plurality of offsets within an M×N block; e) comparing the dispersion metrics corresponding to the plurality of offsets to determine the minimum dispersion coefficient histogram; and f) selecting the offset that corresponds to the minimum dispersion coefficient histogram as the offset for the DCT block boundaries.

Abstract: A method for producing a compressed bit-stream from a digital image includes the steps of a) processing the digital image to produce a main subject belief map containing a continuum of belief values relating to the importance of subject and background regions in the image, b) performing a spatio-frequency transformation on the digital image to produce an array of transform coefficients, c) deriving a distortion-weighting factor for each transform coefficient from the belief map, and d) producing a compressed bit stream using an image compression system that is responsive to the distortion-weighting factors. The specific image compression system may be selected from a variety of image compression systems, including JPEG compression, JPEG2000 compression or vector quantization.

Abstract: A file of digital color data representing a color image is transformed into a new file of digital color data where one or more colors have been transformed to one or more new color locations where their reproduction is known to be preferred. The digital color data is provided in a multi-dimensional color space where the color locations to be operated on in the color space are designated as color magnets. The color space distance is then calculated between the color locations of the digital color data in the color space and the color magnets. A particular activity is prescribed for each color magnet that affects nearby color locations in the color space; this activity includes one or more activity selected from the group including attraction, repulsion, shielding and dragging.

Abstract: A computer vision/image processing method generates a depth map useful in producing a foreground depth mask for 2D/3D image editing. The method uses image data from a plurality of scenes. Feature points on each of the vertical scan lines in each of the scene images are used to search for corresponding feature points on the corresponding vertical lines in other images. The corresponding feature-point search is performed by using a bipartite match network with a feature-point-ordering constraint and a disparity-limit constraint, and produces an individual feature-point depth map for each input image. A sparse feature-point depth map of the scene is obtained after applying a consistency test to all the individual depth maps. A complete feature-point depth map is produced by applying a color property assisted depth propagation process to the sparse feature-point depth map.

Abstract: An image registration method is described for X-ray imagery, in which specific views of the x-rays are provided to optimize tie point selection accuracy and efficiency. The first view maintains context during initial point selection. The second view provides a detailed view of each point pair, to allow for fine adjustment, while automatically presenting each point pair in sequence to the user. After the tie points are refined and the images are registered, a third view is provided which allows direct comparison of the registered images.

Abstract: A method for automatically classifying images into events for composing and authoring of a multimedia image program on a recordable optical disc comprises the steps of: (a) receiving a plurality of images having either or both date and/or time of image capture; (b) determining one or more largest time differences of the plurality of images based on clustering of the images; (c) separating the plurality of images into events based on having one or more boundaries between events which one or more boundaries correspond to the one or more largest time differences; (d) specifying at least one multimedia feature that is related to each event; (e) encoding the images between event boundaries and the at least one multimedia feature associated therewith into an event bitstream; and (f) writing each event bitstream to the recordable optical disc, whereby each event is authored into a separate section of the recordable optical disc.

Abstract: A method is described for enhancing a color digital image by first identifying a local neighborhood of pixels including a pixel of interest and then using the color values of one or more of the pixels from the local neighborhood to calculate a color weighting factor for the pixel of interest, where the color weighting factor is a near continuous function of the location of the pixel of interest in a color space. A spatial filter is applied to the values of pixels from the local neighborhood to calculate a first signal value and a second signal value for the pixel of interest. The color weighting factor is then used to modify either the first value or the second value, and an enhanced pixel value is then generated for the pixel of interest from the first value and second values. The process is then repeated for other pixels in the color digital image.

Abstract: A color scannerless range imaging system comprises four component systems, including an illumination system, an image capture device, an optical assembly and a controller for interconnecting and controlling the component systems. The illumination system separately illuminates a scene with modulated and unmodulated illumination, while the image capture device is positioned in an optical path of the reflected illumination from the scene for capturing a plurality of images thereof, including (a) at least one image of the reflected modulated illumination, whereby the modulation of the reflected modulated illumination incorporates a phase delay corresponding to the distance of objects in the scene from the range imaging system, and (b) at least one color image of the reflected unmodulated illumination.

Abstract: A method for generating a range image sequence from a scannerless range imaging system of the type comprising an illuminator for illuminating a scene with modulated illumination; an image modulating component for receiving and modulating the modulated illumination reflected from the scene; an image capture element for capturing images output by the image modulating component, including an image bundle of at least three phase offset images each incorporating a phase delay corresponding to the distance of objects in the scene from the illuminator, wherein each phase offset image of an image bundle also incorporates a phase offset distinct for each image. According to the method, a sequence of three or more phase offset images are acquired that correspond to the modulated illumination reflected from the scene, whereby the sequence contains overlapping subsequences of successive phase offset images, each subsequence forming an image bundle.

Abstract: An image is automatically assessed with respect to certain features, wherein the assessment is a determination of the degree of importance, interest or attractiveness of the image. First, a digital image is obtained corresponding to the image. Then one or more quantities are computed that are related to one or more features in the digital image, including one or more features pertaining to the content of the digital image. The quantities are processed with a reasoning algorithm that is trained on the opinions of one or more human observers, and an output is obtained from the reasoning algorithm that assesses the image. More specifically, the reasoning algorithm is a Bayesian network that provides a score which, when done for a group of images, selects one image as the emphasis image or the appeal image. The features pertaining to the content of the digital image include people-related features and/or subject-related features.

Abstract: A method for incorporating photographic calibration into digital color management having an abstract color transformation profile includes the steps of: (a) generating an optical reproduction from a photographic film containing an image of a target having a plurality of colored and neutral patches; (b) scanning said film in a transmission scanner and generating film image data; (c) transforming the film image data to data in a first profile connection space using the color management system; (d) reading said optical reproduction and generating reproduction image data; (e) transforming the reproduction image data to data in a second profile connection space using a color management system; and (f) computing an abstract profile to transform said data in the first profile connection space to data in the second profile connection space.

Abstract: A method to automatically vary the compression of images by ranking images within clusters based upon image emphasis. The ranking process computes one or more quantities related to one or more features in each image. The features can include the content of images. The invention processes the quantities with a reasoning algorithm that is trained based on opinions of one or more human observers. The invention applies the quantities to the images to produce the ranking and variably compresses the images depending upon the ranking. The images having a low ranking and are compressed more than images that have a high ranking.

Abstract: A system and method for determining image similarity. The method includes the steps of automatically providing perceptually significant features of main subject or background of a first image; automatically providing perceptually significant features of main subject or background of a second image; automatically comparing the perceptually significant features of the main subject or the background of the first image to the main subject or the background of the second image; and providing an output in response thereto. In the illustrative implementation, the features are provided by a number of belief levels, where the number of belief levels are preferably greater than two. The perceptually significant features include color, texture and/or shape. In the preferred embodiment, the main subject is indicated by a continuously valued belief map.

Abstract: A digital image processing determining the extent of blocking and contouring artifacts in a digital image includes the steps of: determining the extent of blocking artifacts using the column and row difference arrays; determining the extent of contouring artifacts based on an estimated DC quantization step size; and, determining a composite artifact measure as a function of the extent of blocking artifacts and extent of contouring artifacts.

Abstract: A scannerless range imaging system employs a technique for embedding digital data into its image output in a manner that allows exact recovery of its associated images. The range imaging system captures (a) a plurality of phase images of reflected modulated illumination, wherein each image incorporates a phase delay term corresponding to the distance of objects in the scene from the range imaging system, together with a phase offset term unique for each image, and (b) at least one intensity image of reflected unmodulated illumination, and then generates an image bundle of associated images including the plurality of phase images and the intensity image.

Abstract: Reference material in a digital graphics album is specified. Annotation data is extracted from the reference material and may be processed by a natural language processor to produce search keywords. In addition to the keywords, user directives may be provided, both of which are used to conduct a search for related graphical materials. The search is conducted by querying a graphical material database through a network connection. The search results are received and the user can select from the resultant materials for inclusion in the digital graphics album. If no satisfactory material is found, the user can specify a reference graphical image that is processed to produce search criteria that are image content descriptors. The database is again queried in accordance with these descriptors to provide search results for possible inclusion.

Abstract: A method and structure for stabilizing a motion image formed using a sequence of successive frames which includes calculating a motion vector field between adjacent frames; forming a motion vector histogram from horizontal and vertical components of the motion vector field; applying a threshold to the motion vector histogram to produce a thresholded motion vector histogram; generating average horizontal and vertical motion components from the thresholded motion vector histogram; filtering the average horizontal and vertical motion components over a number of frames to identify unwanted horizontal and vertical motion components for each of the frames; and stabilizing the image sequence by shifting each frame according to the corresponding unwanted horizontal and vertical motion components.

Abstract: A method for automatically organizing digitized photographic images into events based on spoken annotations comprises the steps of: providing natural-language text based on spoken annotations corresponding to at least some of the photographic images; extracting predetermined information from the natural-language text that characterizes the annotations of the images; segmenting the images into events by examining each annotation for the presence of certain categories of information which are indicative of a boundary between events; and identifying each event by assembling the categories of information into event descriptions. The invention further comprises the step of summarizing each event by selecting and arranging the event descriptions in a suitable manner, such as in a photographic album.

Abstract: A method of generating captions or semantic labels for an acquired image is based upon similarity between the acquired image and one or more stored images that are maintained in an image database environment, where the stored images have preexisting captions or labels associated with them. The method includes the steps of: a) acquiring an image for evaluation with respect to the stored images; b) automatically extracting metadata from the acquired image without requiring user interaction in the image database environment; c) automatically selecting one or more stored images having metadata similar to the extracted metadata, thereby providing selected stored images with metadata similar to the acquired image; and d) generating captions or labels for the acquired image from the preexisting captions or labels associated with the selected stored images.

Abstract: A method is described for enhancing a digital image channel, e.g., a channel comprising a texture signal, by providing a predetermined estimate of the noise expected for the digital channel based on a predetermined relationship between the image intensity values and the expected noise for given intensities. After a local estimate of signal activity is generated for the digital image channel, a gain adjustment is generated from the predetermined estimate of noise and the local estimate of signal activity; the gain adjustment is applied to the image pixels in the digital channel in order to generate a digital channel with enhanced image values.