Abstract: A system is configured to determine a color distribution of an object moving along a flow direction relative to a spatial filter. The light emanating from the object is time modulated according to the mask features of the spatial filter. First and second detectors are arranged to sense the modulated light. The first detector senses light having a first wavelength spectrum and generates a first electrical output signal in response to the sensed light. The second detector light senses light having a second wavelength spectrum and generates a second electrical output signal in response to the sensed light. Signals from the first and second detectors include information about color distribution of the object.

Abstract: A device includes a spatial filter arranged in a Cartesian coordinate system having orthogonal x, y, and z axes. The spatial filter has mask features that are more light transmissive and mask features that are less light transmissive. The mask features are arranged along the x-axis in the flow direction of a flow path. A detector is positioned to detect light emanating from at least one object in the flow path, the object having a width along the y-axis, a thickness along the z-axis, and a length along the x-axis. Light emanating from the object is time modulated according to the mask features as the object moves along the flow path. The detector is configured to generate a time-varying electrical signal in response to the detected light that includes information about the width or thickness of the object.

Abstract: Spatially modulated light emanating from an object moving along a flow path is used to determine various object characteristics including object length along the flow direction. Light emanating from at least one object moving along in a flow path along a flow direction of a spatial filter is sensed. The intensity of the sensed light is time modulated according to features of the spatial filter. A time varying electrical signal is generated which includes a plurality of pulses in response to the sensed light. Pulse widths of at least some of the pulses are measured at a fraction of a local extremum of the pulses. The length of the object along the flow direction is determined based on the measured pulse widths.

Abstract: One embodiment of the present invention provides a system for tracking and distributing annotations for a video stream. During operation, the system receives, at an annotation server, the video stream originating from a remote field computer, extracts a number of features from the received video stream, and identifies a group of features that matches a known feature group, which is associated with an annotation. The system further associates the identified group of features with the annotation, and forwards the identified group of features and the annotation to the remote field computer, thereby facilitating the remote field computer to associate the annotation with a group of locally extracted features and display the video stream with the annotation placed in a location based at least on locations of the locally extracted features.

Abstract: A system and method of detecting duplicate document content in a large document collection and automatically highlighting duplicate or different document content among the detected document content using two-dimensional visual fingerprints.

Abstract: One embodiment of the present invention provides a system for tracking and distributing annotations for a video stream. During operation, the system receives, at an annotation server, the video stream originating from a remote field computer, extracts a number of features from the received video stream, and identifies a group of features that matches a known feature group, which is associated with an annotation. The system further associates the identified group of features with the annotation, and forwards the identified group of features and the annotation to the remote field computer, thereby facilitating the remote field computer to associate the annotation with a group of locally extracted features and display the video stream with the annotation placed in a location based at least on locations of the locally extracted features.

Abstract: One embodiment of the present invention provides a system for sharing annotated videos. During operation, the system establishes a real-time video-sharing session between a remote field computer and a local computer. During the established real-time video-sharing session, the system receives a real-time video stream from a remote field computer, forwards the real-time video stream to a local computer to allow an expert to provide an annotation to the real-time video stream, receives the annotation from the local computer, and forwards the annotation to the remote field computer, which associates the annotation with a corresponding portion of the real-time video stream and displays the annotation on top of the corresponding portion of the real-time video stream.

Abstract: Visual fingerprinting is used to provide a robust and highly effective method of finding similar content in a large document collection of rich document content composed of multiple text, line-art, and photo image objects. The visual fingerprints capture unique two-dimensional localized aspects of document appearance. The visual fingerprints are highly distinctive; fast for lookup; compact for storage requirements; and scalable to large document collections.

Abstract: A method for electronically magnifying a target object with an imaging device. The method includes obtaining a full view image of the target object, where the full view image is focused. The method further includes moving the imaging device in proximity to a portion of the target object and obtaining a key image of the portion of the target object. The method further includes matching the key image to a corresponding portion of the full view image. The method further includes magnifying the corresponding portion of the full view image and displaying the magnified portion of the full view image.

Abstract: A method and system is provided for finding stable keypoints in a picture image using localized scale properties. An integral image of an input image is calculated. Then a scale space pyramid layer representation of the input image is constructed at mulitple scales, wherein at each scale, a set of specific filters are applied to the input image to produce an approximation of at least a portion of the input image. Outputs from filters are combined together to form a single function of scale and space. Stable keypoint locations are identified in each scale at pixel locations at which the single function attains a local peak value. The stable keypoint locations which have been identified are then stored in a memory storage.

Abstract: An image management method and system provides for storing, indexing, searching, and/or retrieving image data. Keypoints are identified in images, including keypoints in a query image of a query document, and keypoints in potential target document images of a collection of potential target documents. Fingerprint information from the keypoints are generated, and the fingerprint information of a query image is compared with fingerprint information of potential target document images, found in the collection of potential target documents. A best match is determined between the fingerprint information of the query image and the potential target document images. At least one target document image is retrieved based on the determined best match. The retrieved at least one target image may then be displayed, printed or transmitted.

Abstract: Visual fingerprinting is used to provide a robust and highly effective method of finding similar content in a large document collection of rich document content composed of multiple text, line-art, and photo image objects. The visual fingerprints capture unique two-dimensional localized aspects of document appearance. The visual fingerprints are highly distinctive; fast for lookup; compact for storage requirements; and scalable to large document collections.

Abstract: A system and method of detecting duplicate document content in a large document collection and automatically highlighting duplicate or different document content among the detected document content using two-dimensional visual fingerprints.

Abstract: An image management method and system provides for storing, indexing, searching and/or retrieving image data. The content of an image collection is pre-processed to identify stable and repeatable keypoints for each image in the collection. Fingerprint information is computed from local groups of keypoints, and the resulting fingerprint information is stored in a compact fingerprint database. The computing of the fingerprint information is based on combinations of a subgroup of the strongest keypoints in an image, called anchor keypoints, in addition to a number of non-anchor keypoints. For each fingerprint in the compact fingerprint database, a sequence of candidate fingerprint combinations is generated and stored in a Fan Tree and/or hashtable and a corresponding fingerprint data structure. The sequence of fingerprint combinations allows several non-anchor keypoints to be missing, while still allowing the system and method to correctly detect fingerprints with high accuracy.

Abstract: One embodiment of the present invention provides a system that facilitates interaction between two entities located away from each other. The system includes a virtual reality system, an augmented reality system, and an object-state-maintaining mechanism. During operation, the virtual reality system displays an object associated with a real-world object. The augmented reality system displays the object based on a change to the state of the object. The object-state-maintaining mechanism determines the state of the object and communicates a state change to the virtual reality system, the augmented reality system, or both. A respective state change of the object can be based on one or more of: a state change of the real-world object; a user input to the virtual reality system or the augmented reality system; and an analysis of an image of the real-world object.

Abstract: A method and system generates fine-grained fingerprints for identifying content in a rendered document. It includes applying image-based techniques to identify patterns in a document rendered by an electronic document rendering system, irrespective of a file format in which the rendered document was electronically created. The applying of the image-based technique includes identifying candidate keypoints at locations in a local image neighborhood of the document, and combining the locations of the candidate keypoints to form a fine-grained fingerprint identifying patterns representing content in the document.

Abstract: A method and system detects and highlights changes in documents and displays those documents in a side-by-side aligned view. Aspects of a source document and a revised document are detected and compared. Similarities and/or differences between the source document and the revised document are identified and visual identifiers are introduced to maintain consistent and accurate alignment between content in the source document and content in the revised document. A merged single file is output containing the aligned side-by-side view of the source document and the revised document, with all differences between the source document and the revised document visually identified.

Abstract: A method and system generates fine-grained fingerprints for identifying content in a rendered document. It includes applying image-based techniques to identify patterns in a document rendered by an electronic document rendering system, irrespective of a file format in which the rendered document was electronically created. The applying of the image-based technique includes identifying candidate keypoints at locations in a local image neighborhood of the document, and combining the locations of the candidate keypoints to form a fine-grained fingerprint identifying patterns representing content in the document.

Abstract: A method for electronically magnifying a target object with an imaging device. The method includes obtaining a full view image of the target object, where the full view image is focused. The method further includes moving the imaging device in proximity to a portion of the target object and obtaining a key image of the portion of the target object. The method further includes matching the key image to a corresponding portion of the full view image. The method further includes magnifying the corresponding portion of the full view image and displaying the magnified portion of the full view image.

Abstract: A system and method for tetrahedral interpolation computations using data-level parallelism that takes advantage of data-level parallelism in media processors. If the tetrahedron points in a 3D lookup table are packed together in a memory, the interpolation computation can be implemented without extra instructions to unpack them. An algebraic manipulation of the interpolation equation allows computing the difference on the fraction coefficients instead of the tetrahedron node values. Not only will this technique preserve the full precision without over or underflow, but the packed data from the 3D lookup can be used directly, thereby allowing a faster implementation of the color space transformation overall and implementing as part of a direct-copy image path on a media processor. Such a system and method allows the implementation of the direct copy pipeline to function at the required performance rate as requested by a customer specification while obtaining the required product design speed.