Abstract: Disclosed are various embodiments for tracking an object shown as moving in a video. One embodiment is a method for tracking an object in a video that comprises tracking in a first temporal direction an object in a plurality of video frames and generating a first tracking result, evaluating the first tracking result corresponding to tracking of the object in the first temporal direction, and stopping tracking in the first temporal direction upon the occurrence of a predefined event, wherein the predefined event is based on an evaluated tracking result. The method further comprises obtaining data identifying an object outline of the object upon stopping the tracking in the first temporal direction, tracking in a second temporal direction the object based on the data identifying the object outline of the object to generate a second tracking result, and generating a refined tracking result based on at least on one of the first tracking result, the second tracking result, or a combination thereof.

Abstract: A vehicle periphery monitoring device includes: a first edge image generation element 5 which generates a first edge image on the basis of luminance components of a captured image acquired by an in-vehicle camera 2; a second edge image generation element 6 which generates a second edge image on the basis of hue components or saturation components of the captured image; a composite edge image generation element 7 which generates a composite edge image formed by combining the first edge image and the second edge image; and an object classification identification element 8 which identifies whether or not the object is a prescribed kind of structure on the basis of the external shape of the object represented by the composite edge image.

Abstract: Mechanisms are provided for determining the physical location of a physical asset in a physical area. A plurality of physical assets are controlled to cause each physical asset to output a visual output pattern on visual output elements of the physical asset. An image of a target physical asset is captured that has the current state of the visual output elements. An identification of the target physical asset is determined based on the current state of the visual output elements. A physical location of the target physical asset is determined based on a physical location of the image capture device when the image was captured. Location data identifying the determined physical location of the target physical asset is stored in an asset database in association with configuration information for the physical asset.

Abstract: Provided are an image processing apparatus, an image processing method, a computer-readable medium storing a computer program and an image processing system improving edge detection precision at the time of flare occurrence. The image processing apparatus includes an input unit for taking image data including a document region, a first candidate detector for detecting a first candidate of an edge point constituting a boundary line of the document region by scanning binarized image of the image data with a predetermined pattern along a line, a second candidate detector for detecting a second candidate of an edge point based on differential value of pixels adjoining each other, and an edge point determination unit for determining the second candidate as an edge point when the second candidate is positioned more inside of the document region than the first candidate, and otherwise determines the first candidate as an edge point.

Abstract: A processing-target image generation device generates a processing-target image which is an object to be subjected to an image conversion process for acquiring an output image based on an input image taken by an image-taking part. A coordinates correspondence part causes input coordinates, spatial coordinates, and projection coordinates to correspond to each other, the input coordinates being on an input image plane on which the input image is located, the spatial coordinates being on a space model on which the input image is projected, the projection coordinates being on a processing-target image plane on which the processing-target image is positioned and the image projected on the space model is re-projected.

Abstract: The invention relates to a method and an apparatus for processing a workpiece, such as a wall, a ceiling, a floor or the like in a building, by means of an appliance which can be moved, in particular an electric tool, wherein the appliance can be moved to at least one predeterminable nominal position. Essentially continuous position identification, in particular by means of optical methods, is carried out for the respective actual position of the appliance and/or of an object which can be moved corresponding to the appliance. The position information, such as the coordinates of the respective determined actual position and/or information relating to any, discrepancy between the nominal position and the actual position, is transmitted wirelessly to the appliance. The respective position information is indicated to the user on the appliance, for manual position control.

Abstract: An image processing apparatus includes: an image receiver which receives a predetermined image obtained by photographing a fetus; and a controller which detects a head region and a torso region of the fetus from the predetermined image, and which models a shape of the fetus by using at least one of a first contoured shape corresponding to the detected head region, a second contoured shape corresponding to the detected torso region, a first axis that is the central axis of the detected head region, and a second axis that is the central axis of the detected torso region, to model the fetus so that biometric data of the fetus can be easily measured.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: The method, system and computer program product generate online interactive maps linking tracked objects (in a live or pre-recorded video sequence) to multimedia content in real time. Specifically, an object tracking and link generation (OTLG) utility allows a user to access multimedia content by clicking on moving (or still) objects within the frames of a video (or image) sequence. The OTLG utility identifies and stores a clear image(s) of an object or of multiple objects to be tracked and initiates a mechanism to track the identified objects over a sequence of video or image frames. The OTLG utility utilizes the results of the tracking mechanism to generate, for each video frame, an interactive map frame with interactive links placed in the map frame at a location corresponding to the object's tracked location in each video frame.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: The present invention provides an image processing apparatus for performing image processing of image data in which information specifying, on a first coordinate system, a position of a portion of interest in an image is recorded, including an image processing unit configured to perform rotation processing of the image data using a second coordinate system having an origin different from that of the first coordinate system, and a change unit configured to change the information specifying the position of the portion of interest in accordance with a rotation amount of the image data by the rotation processing such that an image of the portion of interest specified by the information specifying the position of the portion of interest after the rotation processing matches that before the rotation processing.

Abstract: Method for tracking entities (2) evolving in a monitored space having a metrics with at least one predetermined granularity, said method including repetition of the following operations: determining a current spatial and/or temporal state of the entities (2) within said space, according to said metrics, checking occurrence of a triggering event in connection with said state, upon occurrence of said triggering event, changing granularity of the monitored space for at least one entity (2).

Abstract: A media object, such as an image file, a video file, or an audio file, is analyzed to determine relationships between persons associated with the media object, which may include persons captured in the media object and/or a person that captured the media object. A representation of a first person captured in a media object is detected. The media object is analyzed to determine at least one indicator of a relation between the first person and a second person associated with the media object. A relationship between the first person and the second person is predicted based at least on the determined at least one relation indicator. The media object may be monetized in various ways, such as by directing advertisements to persons associated with the media object, and/or to persons having social connections to the persons associated with the media object.

Abstract: Detecting an object includes receiving depth data and infrared (IR) data from a depth sensor. A first background subtraction is performed on the IR data to create a first mask, and a second background subtraction is performed on the IR data to create a second mask. The first and second masks and the depth data are merged to create a third mask.

Abstract: A device is provided with a luminance histogram calculation unit which generates a luminance histogram showing appearance frequency of luminance values contained within the infrared image and determines a luminance value corresponding to a peak in the luminance histogram as a background luminance level of the background; a luminance shift calculation unit which sets the background luminance value as an intermediate value in luminance range width of the infrared image and generates a luminance shift image by linearly shifting other luminance values in the infrared image based on the intermediate value; a reversed image processing unit which generates a reversed shift image wherein the luminance level of the luminance shift image is reversed; and a luminance calculation processing unit which generates a calculation-processed image by performing calculation processing based on the difference in the luminance values at corresponding positions in the luminance shift image and the reversed shift image.

Abstract: A computer implemented method for evaluating a one-to-one mapping between a first spatial point set and a second spatial point set in nD comprising the steps of receiving a first and a second spatial point sets in nD and a one-to-one mapping between the two spatial point sets; generating a pair of mapped agreeable (n+1)-combinations in the first point set; computing two affine transformations that transform the pair of mapped agreeable (n+1)-combinations to correspondents in the second point set; computing the difference of the left sub-matrices of the two affine transformations; and computing a local distance measure based on the difference of the left sub-matrices of the two affine transformations.

Abstract: Apparatus and method for processing a sequence of images of a scene, the method including: tracking a region of interest in the sequence of images (e.g. using a Self Adaptive Discriminant filter), selecting a particular image in the sequence, selecting a set of images from the sequence, the set of images including one or more images that precede the particular image in the sequence of images; and determining a value indicative of the level of change between the region of interest in the particular image and the regions of interest in the images in the set of images (e.g. using a Change Detection Process).

Abstract: A method and system for people flow statistics is disclosed in the invention, wherein said method comprises: multi-types of classifiers connected in parallel are used to perform the human head detection in the current image, the respective human heads in the current image are determined; the respective determined human heads are tracked to form the human head target movement tracks; and the people flow is counted at the direction of the human head target movement track. It can be seen, a plurality of classifiers connected in parallel are used in the invention, multi-types of human head targets, such as dark colored hair, light colored hair and caps of various colors, and the like, can be detected simultaneously.

Abstract: Apparatus and method for processing a sequence of images of a scene, the method including: tracking a region of interest in the sequence of images (e.g. using a Self Adaptive Discriminant filter); selecting a particular image in the sequence; selecting a set of images from the sequence, the set having one or more images that precede the particular image in the sequence of images; for each pixel in the region of interest in the particular image, determining a value for a parameter; for each pixel in the region of interest of each image in the set of images, determining a value for the parameters; and comparing a function of the determined values for the region of interest in the particular image to a further function of the determined values for the regions of interest in the images in the set of images.

Abstract: An obstacle detection display device, for enabling to display the information of a result of image recognition at high accuracy, at high speed and without position error of shift, even for an image photographed by a wide-angle camera, is constructed with: an image pickup means; an image correcting means for executing geometrical correction upon a video signal outputted from the image pickup means through image processing, and thereby outputting it; an image recognizing means for detecting object information by executing image processing on the video signal outputted from the image correcting means; a display information producing means for determining display information and display position in the video signal, to be presented to a user, being superimposed on the video signal, which is outputted by the image pickup means, upon basis of the object information detected by the image recognizing means, and thereby outputting them; and a display information superimposing means for superimposing the display inform

Abstract: An inspection apparatus, inspection system, inspection method, and inspection control program stored in a recording medium, each of which sets a reference point in a master image and a read image read from a printed image, which is to be used for detecting the positional shift between the read image and the master image, based on determination whether a pattern previously added to the printed image for detecting the positional shift is available or can be used to effectively detect the positional shift.

Abstract: Determining the position and orientation of base station antennas in mobile radio is highly important for optimizing the transmission parameters. It requires time-consuming and costly effort, since measurements need to be taken directly at the antenna. Furthermore, it is often impossible to access the antennas. This has a very disadvantageous effect on the precision of the data, which means that the optimizing results suffer greatly from this. The method according to the invention and the presented apparatus allow the position and orientation of the transmission antenna to be found quickly, precisely and inexpensively without needing to take measurements close to the antenna. This involves the absolute position of the antenna being determined by means of automatic determination of the coordinates of the reference point, measurement of distance to the target antenna, measurement of the angle of inclination and measurement of the orientation of the measuring equipment.

Abstract: Each image of a set of images is characterized with a set of sparse feature descriptors and a set of dense feature descriptors. In some embodiments, both the set of sparse feature descriptors and the set of dense feature descriptors are calculated based on a fixed rotation for computing texture descriptors, while color descriptors are rotation invariant. In some embodiments, the descriptors of both sparse and dense features are then quantized into visual words. Each database image is represented by a feature index including the visual words computed from both sparse and dense features. A query image is characterized with the visual words computed from both sparse and dense features of the query image. A rotated local Bag-of-Features (BoF) operation is performed upon a set of rotated query images against the set of database images. Each of the set of images is ranked based on the rotated local Bag-of-Features operation.

Abstract: A method of data processing is provided for estimating a position of an object in an image from a position of a reference object in a reference image. The method includes learning the position of the reference object in the reference image and its relation to a set of reference landmarks in the reference image, accessing the image, accessing the relation between the position of the reference object and the set of the reference landmarks, identifying a set of landmarks in the image corresponding to the set of the reference landmarks, and applying the relation to the set of landmarks in the image for estimating the position of the object in the image.

Abstract: Techniques and tools are described for performing perspective correction using a reflection. Reflective properties of a surface being photographed can be used to determine a rotation of the device taking the photograph relative to the surface. Light sourced or produced by the device can be used to create a reflection spot in the picture. A position of the reflection spot within the picture is calculated and used to determine the rotation. The rotation can be used for performing perspective correction on the picture, or on another picture taken by the device.

Abstract: The present invention provides a method, apparatus and system of content creation, content post processing, new image capture and image composition whereby alternate images or video content may be selectively integrated into an original video in real time. In the present invention, sufficient metadata is created at the time of the creation of the original video content to both describe objects/areas in the original video that are available for integration of alternate images/video and to provide a means of controlling the alternate video capture and processing to fit the alternate image/video into the areas in the original video defined as available for integration of alternate image/video. The result is a mechanism whereby personal/alternate video/images may be integrated with existing video content, such as in one embodiment, music videos (e.g., video Karaoke), games, broadcast video, DVD videos, and the like, in real time.

Abstract: Systems, methods, and machine readable and executable instructions are provided for hand gesture recognition. A method for hand gesture recognition can include detecting, with an image input device in communication with a computing device, movement of an object. A hand pose associated with the moving object is recognized and a response corresponding to the hand pose is initiated.

Abstract: An information interchange unit, a storage unit, and a display controller are configured such that, after a image selection unit selects a first image and a second image, the information interchange unit interchanges, automatically, first image information of the first image with second image information of the second image, or interchanges, automatically, first position information of the first image with second position information of the second image, the storage unit stores and correlates the first image information and the second position information, and stores and correlates the second image information and the first position information, and the display controller controls, automatically, a display to display the one image based on the first image information and the second position information, and the another image based on the second image information and the first position information.

Abstract: A process and system that allows an end user to insert images into a video content so that the images appear part of that content. This allows a consumer to seem to be part of a video, and interact with the famous personalities in that video. A video content is produced and a first software program is used to insert place holding templates into the video content. The end user produces a video image satisfying the requirements of the inserted template, and a second software program composites the two video streams so that the end user's video image is inserted into the first video content replacing the templates. The system can also be used to insert advertising information into a background to create custom, personalized, “on the fly” advertisements and product placements.

Abstract: Multi view multi spectral detection and tracking system comprising at lease one imager, at least one of the at least one imager being a multi spectral imager, the at least one imager acquiring at least two detection sequences, and at least two tracking sequences, each sequence including at least one image, each acquired image being associated with respective image attributes, an object detector, coupled with the at least one imager, detecting objects of interest in the scene, according to the detection sequence of images and the respective image attributes, an object tracker coupled with the object detector, the object tracker tracking the objects of interest in the scene and determining dynamic spatial characteristics and dynamic spectral characteristics for each object of interest according to the tracking sequences of images and the respective image attributes and an object classifier, coupled with the object tracker, classifying the objects of interest according to the dynamic spatial characteristics and

Abstract: An obstacle region candidate point relating unit assumes that a pixel in an image corresponds to a point on a road surface, and associates pixels between images at two times on the basis of the amount of movement of a vehicle in question, a road plane, and a flow of the image estimated. When a pixel corresponds to a shadow of the vehicle in question or the moving object therearound appearing on the road surface, the ratio of intensities of the pixel values of the spectral images between two images should be approximately the same as the ratio of the spectral characteristics of the sunshine in the sun and the shade. Therefore, when the ratio of intensities is approximately the same as the ratio of the spectral characteristics, the obstacle determining unit does not determine that the pixel in question is a point corresponding to the obstacle.

Abstract: A hand gesture from a camera input is detected using an image processing module of a consumer electronics device. The detected hand gesture is identified from a vocabulary of hand gestures. The electronics device is controlled in response to the identified hand gesture. This abstract is not to be considered limiting, since other embodiments may deviate from the features described in this abstract.

Abstract: A method and a system for determining a parking boundary violation includes receiving video data from a sequence of frames taken from an image capture device monitoring a parking area including at least a single parking space. A boundary is determined for defining at least one parking space in the parking area. A vehicle is detected in the parking area. A determination is made whether the detected vehicle is violating a parking regulation based on a position and size of the vehicle within the boundary.

Abstract: There is provided an image processing device including: a data storage unit that stores object identification data for identifying an object operable by a user and feature data indicating a feature of appearance of each object; an environment map storage unit that stores an environment map representing a position of one or more objects existing in a real space and generated based on an input image obtained by imaging the real space using an imaging device and the feature data stored in the data storage unit; and a selecting unit that selects at least one object recognized as being operable based on the object identification data, out of the objects included in the environment map stored in the environment map storage unit, as a candidate object being a possible operation target by a user.

Abstract: A method and a system for determining a parking angle violation includes receiving video data from a sequence of frames taken from an image capture device monitoring a parking area. The method further includes determining a first line in a current frame where the line represents a nominal orientation of the parking area. The method includes detecting a presence of a vehicle in the parking area. The method includes determining a second line in the frame where the line represents the orientation of the detected vehicle. The method further includes computing an angle between the first and second lines. The method includes determining whether the detected vehicle is violating a parking regulation based on the computed angle.

Abstract: A method of matching fingerprints is disclosed. The method comprises, for a first minutia point, being assigned as a planet minutia point a) determining pairs including the planet minutia point and a satellite minutia point, respectively, such that a cluster is formed; b) comparing the clusters of the respective sets and excluding non-matching satellites; c) counting links in the cluster formed by remaining pairs; and d) for remaining satellite minutia points, performing steps a) to c) with respective satellite minutia point assigned as planet minutia point to form a supercluster by iterating steps a) to d) and superadding the clusters; calculating a score of the supercluster based on the aggregate counted links; and comparing the score with a threshold. A biometric matching apparatus, a portable data carrier, a data processing unit comprising a matching apparatus and a computer program for implementing the invention are also disclosed.

Abstract: An inspection apparatus determines information indicative of structure that may be hidden behind an obscuring boundary, such as a wall. A processor collects measurements of properties characterizing the hidden structure and measurements of location of the apparatus. The collected data are mapped to produce an image of intensity in the characteristic measurements. Each intensity value in the image reflects a measure of density, of material type, or of some other specific information by which hidden structure can be discerned. The intensity changes indicating the hidden structure are displayed to a user via color-coded pixels or the like.

Abstract: An optics detection system is disclosed. The optics detection system includes a sensor module and a processor. The sensor module is configured to illuminate a field of regard with a plurality of light pulses and to capture reflections of the plurality of light pulses in a plurality of frames, respectively. The processor is configured to process the plurality of frames to locate and identify optics within the field of regard using a plurality of discriminators.

Abstract: A game apparatus detects a predetermined image object including a first graphic pattern with a plurality of inner graphic patterns drawn therein from a captured image captured by an image-capturing section. The game apparatus first obtains the captured image captured by the image-capturing section, and detects an area of the first graphic pattern from the captured image. Then, the game apparatus detects the plurality of inner graphic patterns from within the detected area, and calculates center positions of the inner graphic patterns so as to detect the position of the predetermined image object.

Abstract: This specification relates to locating objects using indicia. In general, one innovative aspect of the subject matter described in this specification can be embodied in methods that include the actions of obtaining an image captured from a camera in proximity of an object located in an indoor facility, the image being of a portion of a surface of the indoor facility comprising a plurality of visible indicia, the camera having an orientation generally pointed at the surface. A plurality of local indicia are identified within the image. The locations of the local indicia within the image and an index of the visible indicia of the surface are used to determine the location of the object relative to the surface.

Abstract: The position of a feature within an image is determined by determining an initial estimate of the feature position to within a fraction of a pixel, translating the feature by an amount equal to that fraction of a pixel, determining a further estimate of the translated feature position to within a fraction of a pixel and summing the pixel fraction of the previous estimate with the further estimate of the position to arrive at a refined estimate of the feature position.

Abstract: A method and system for occlusion region detection and measurement between a pair of images are disclosed. A processing device receives a first image and a second image. The processing device estimates a field of motion vectors between the first image and the second image. The processing device motion compensates the first image toward the second image to obtain a motion-compensated image. The processing device compares a plurality of pixel values of the motion-compensated image to a plurality of pixels of the first image to estimate an error field. The processing device inputs the error field to a weighted error cost function to obtain an initial occlusion map. The processing device regularizes the initial occlusion map to obtain a regularized occlusion map.

Abstract: A method for determining processed image data about a surround field of a vehicle includes a step of combining a map information item with a position and an orientation of the vehicle and with an image of the surround field acquired by a surround-field monitoring device of the vehicle, in order to obtain a position-specific image of the surround field. In addition, the method includes a step of processing the position-specific image so as to include at least one symbol that is assigned to a classified object in the position-specific image, in order to determine the processed image data.

Abstract: An information processing apparatus includes a selection unit configured to select a plurality of specific areas of a target object, a learning unit configured to learn a detection model that relates to each of the plurality of specific areas, a generation unit configured to generate an area combination as a combination of specific areas selected from the plurality of specific areas, a recognition unit configured to recognize the target object based on the detection model and the area combination, and an addition unit configured to add a new specific area based on a recognition result obtained by the recognition unit. If the new specific area is added by the addition unit, the learning unit further learns a detection model that relates to the new specific area.

Abstract: Provided are an image processing apparatus and method for counting moving objects in an image, the apparatus including: a motion detection unit which detects motion in an image; an object detection unit which detects objects based on the motion detected by the motion detection unit; an outline generation unit which generates at least one reference outline of which a size is adjusted according to a preset parameter based on a location in the image; and a calculation unit which calculates a number of objects having substantially a same size as that of the at least one reference outline from among the objects detected by the object detection unit, wherein the preset parameter is adjusted according to at least one circumstantial parameter.

Abstract: Systems and methods are disclosed to perform object detection for images from an image sensor by reusing a 1-dimensional feature from a previously learned weak classifier and selecting a new feature to construct a 2-dimensional feature space; and cross partitioning the 2-dimensional space to learn optimal outputs for instances in each domain within a boosting framework.

Abstract: A method is provided for determining the position of a first structure (8a) relative to a second structure (8b) or a part thereof, said method having the steps of: a) providing a first picture (F1) having a multiplicity of pixels and which contains the first structure, b) providing a second picture (F2) having a multiplicity of pixels and which contains the second structure, c) forming an optimization function with the displacement of the two pictures relative to one another as parameter, the optimization function overlying the two pictures and masking the overlay such that in a determination of an extreme value of the optimization function a contribution is made only by the region of the overlay that corresponds to the second structure or the part thereof, d) ascertaining the extreme value of the optimization function and determining the optimal value of the displacement based on the extreme value of the optimization function, and e) determining the position of the first structure relative to the second stru

Abstract: Techniques are provided for a gesture device to detect a series of gestures performed by a user and execute corresponding electronic commands associated with the gestures. The gesture device detects a gesture constituting movements from a user in three-dimensional space and generates a confidence score value for the gesture. The gesture device selects an electronic command associated with the gesture and compares the electronic command with a prior electronic command associated with a prior gesture previously detected by the gesture device in order to determine a compatibility metric between the electronic command and the prior electronic command. The gesture device then adjusts the confidence score value based on the compatibility metric to obtain a modified confidence score value. The electronic command is executed by the gesture device when the modified confidence score value is greater than a predetermined threshold confidence score value.

Abstract: A method and a system for identifying a vehicle in a parking area includes receiving video data from a video camera observing a parking area from a first field of view over duration of time and receiving image data from a still camera observing a second field of view overlapping the first field of view. The method includes tracking a location of a vehicle across a sequence of frames. The tracking includes determining a time instant at which each frame was captured. The method includes comparing a select frame captured by the video camera with an image captured by the still camera. Based on the comparing, the method includes matching the vehicle in the image with the vehicle in the frame. The method includes determining a license plate number of the vehicle by locating a license plate on the vehicle in the image.