Abstract: An example disclosed apparatus includes memory configured to store a mapping of items to respective ones of a plurality of minimum pixel-resolutions, wherein the minimum pixel-resolutions correspond to an image resolution required to identify a corresponding one of the items using an image; and a processor configured to obtain identifying information corresponding to one of the items; select, using the mapping and based on the identifying information, one of the plurality of minimum pixel-resolutions; determine a value of an output parameter based on the selected one of the plurality of minimum pixel resolutions and a camera position relative to a shelf, wherein the output parameter is at least one of an elevation angle of the camera, a vertical resolution, a horizontal resolution, a vertical field of view, and a horizontal field of view; and output the determined value of the output parameter to facilitate positioning of the camera.

Abstract: An example disclosed apparatus includes memory configured to store a mapping of items to respective ones of a plurality of minimum pixel-resolutions, wherein the minimum pixel-resolutions correspond to an image resolution required to identify a corresponding one of the items using an image; and a processor configured to obtain identifying information corresponding to one of the items; select, using the mapping and based on the identifying information, one of the plurality of minimum pixel-resolutions; determine a value of an output parameter based on the selected one of the plurality of minimum pixel resolutions and a camera position relative to a shelf, wherein the output parameter is at least one of an elevation angle of the camera, a vertical resolution, a horizontal resolution, a vertical field of view, and a horizontal field of view; and output the determined value of the output parameter to facilitate positioning of the camera.

Abstract: An apparatus including a display screen displaying input fields for accepting user-input a list of input parameters that includes a height of the camera, a distance from the camera to the shelf, at least one of a width-span and a height-span. The apparatus also includes a memory to store a required minimum pixel-resolution of a selected label and to store a list of camera types, and includes a microprocessor configured to determine at least an elevation angle from the list of input parameters. The display screen can also be configured for displaying at least one camera type selected from the list of camera types based a selection criterion that depends upon at least some of said output parameters.

Abstract: An example disclosed method includes receiving an image containing shelf edge content displayed on a shelf edge; determining a location of the shelf edge content and an apparent orientation of the shelf edge content; determining, based on the location of the shelf edge content and the apparent orientation of the shelf edge content, a region in the image corresponding to the shelf edge; and providing an output based on the image that identifies the region in the image corresponding to the shelf edge.

Abstract: An example disclosed method includes receiving an image containing shelf edge content displayed on a shelf edge; determining a location of the shelf edge content and an apparent orientation of the shelf edge content; determining, based on the location of the shelf edge content and the apparent orientation of the shelf edge content, a region in the image corresponding to the shelf edge; and providing an output based on the image that identifies the region in the image corresponding to the shelf edge.

Abstract: A method and apparatus for image processing to avoid counting shelf edge promotional labels when counting product labels. Shelf edges are identified by detecting shelf edge content in a captured image by comparing the image to reference images of shelf edge content. Detected occurrences of shelf edge content are demarcated using a geometric pattern having corners at coordinates corresponding to positions around the identified shelf edge content. Corresponding corners are grouped into clusters, and the clusters are analyzed to define the upper and lower bounds of a shelf region in the image.

Abstract: An apparatus including a display screen displaying input fields for accepting user-input a list of input parameters that includes a height of the camera, a distance from the camera to the shelf, at least one of a width-span and a height-span. The apparatus also includes a memory to store a required minimum pixel-resolution of a selected label and to store a list of camera types, and includes a microprocessor configured to determine at least an elevation angle from the list of input parameters. The display screen can also be configured for displaying at least one camera type selected from the list of camera types based a selection criterion that depends upon at least some of said output parameters.

Abstract: An improved object recognition method is provided that enables the recognition of many objects in a single image. Multiple instances of an object in an image can now be detected with high accuracy. The method receives a plurality of matches of feature points between a database image and a query image and determines a kernel bandwidth based on statistics of the database image. The kernel bandwidth is used in clustering the matches. The clustered matches are then analyzed to determine the number of instances of the object within each cluster. A recursive geometric fitting can be applied to each cluster to further improve accuracy.

Abstract: A method and apparatus for image processing to avoid counting shelf edge promotional labels when counting product labels. Shelf edges are identified by detecting shelf edge content in a captured image by comparing the image to reference images of shelf edge content. Detected occurrences of shelf edge content are demarcated using a geometric pattern having corners at coordinates corresponding to positions around the identified shelf edge content. Corresponding corners are grouped into clusters, and the clusters are analyzed to define the upper and lower bounds of a shelf region in the image.

Abstract: A video analytic device performs a method for detecting people within frames of video based upon multiple colors within their clothing. The method includes: receiving a frame of video; and determining that a first color region within the frame matches a first color of interest for a clothing uniform, wherein the determining is based on a first set of color representation constraints. The method further includes determining that a second color region within the frame matches a second color of interest for the clothing uniform, wherein the determining is based on a second set of color representation constraints and the first and second colors of interest are different. In addition, the method includes applying a set of geometric constraints to the first and second color regions to determine a count of people within the frame wearing the clothing uniform.

Abstract: An improved object recognition method is provided that enables the recognition of many objects in a single image. Multiple instances of an object in an image can now be detected with high accuracy. The method receives a plurality of matches of feature points between a database image and a query image and determines a kernel bandwidth based on statistics of the database image. The kernel bandwidth is used in clustering the matches. The clustered matches are then analyzed to determine the number of instances of the object within each cluster. A recursive geometric fitting can be applied to each cluster to further improve accuracy.

Abstract: A method and apparatus for providing targeted advertisements is provided herein. In particular, targeted advertisements are provided to users based on face clustering for time-varying video. During operation video is continuously obtained of users of the system. Users' faces are detected and measured. Measurements of users' faces are then clustered. Once the clusters are available, advertisements are targeted at clusters rather than individual users.

Abstract: A method and apparatus for selective transmission of color information to a remote recognition system is provided herein. During operation, a central server may determine that at least one region in an image is ambiguously recognized, or unrecognized. In response, the central server will send a request for the image/video to be provided in color, or alternatively for a portion of the image/video to be provided in color. Because only selective images, or portions of images will be transmitted in color, the above transmission scheme enables reduction of bandwidth, and hence cost, required for transmitting images to the analytics server, without compromising the accuracy of the analytics.

Abstract: A method of providing three dimensional (3D) information is disclosed in which a master recruits a local slave to provide contemporaneous images. The images are collected and 3D information extracted therefrom. The 3D information or various images may be transmitted to a remote device and is employed to provide a substitute background that replaces the background of the original image. Either the new image is displayed on one or both devices. The images can be collected and the 3D information extracted at either of the master or slave or at a separate device. To recruit the slave, the master broadcasts a query to all local devices, determines the available devices and their characteristics, automatically or manually selects a particular device, requests authorization from the selected device, and pairs with the selected device when authorization is received. Devices of the same or different types may be paired.

Abstract: A system and method for reducing distance-based distortion in a camera image of an object, where the distanced-based distortion is due to differences in distance from the camera to different parts of the object. In one approach, the distortion is reduced by estimating distances to different parts of the object and then generating a reprojected image of the object dependent upon the estimated distances and upon a virtual viewpoint that is more distant than the camera from the object. In a further approach, the image is warped such that points in the image match corresponding points in one or more stored templates. In a still further approach, if excessive distortion is present in the image, the camera zoom is increased and a magnified image is displayed, prompting a person to move farther from the camera thereby reducing the distortion.

Abstract: A video analytic device performs a method for detecting people within frames of video based upon multiple colors within their clothing. The method includes: receiving a frame of video; and determining that a first color region within the frame matches a first color of interest for a clothing uniform, wherein the determining is based on a first set of color representation constraints. The method further includes determining that a second color region within the frame matches a second color of interest for the clothing uniform, wherein the determining is based on a second set of color representation constraints and the first and second colors of interest are different. In addition, the method includes applying a set of geometric constraints to the first and second color regions to determine a count of people within the frame wearing the clothing uniform.

Abstract: A method and apparatus for providing targeted advertisements is provided herein. In particular, targeted advertisements are provided to users based on face clustering for time-varying video. During operation video is continuously obtained of users of the system. Users' faces are detected and measured. Measurements of users' faces are then clustered. Once the clusters are available, advertisements are targeted at clusters rather than individual users.

Abstract: A method is performed for selecting a video analysis method based on available video representation features. The method includes: determining a plurality of available video representation features for a first video output from a first video source and for a second video output from a second video source; and analyzing the plurality of video representation features as compared to at least one threshold to select one of a plurality of video analysis methods to track an object between the first and the second videos.

Abstract: The processing of a stereoscopic image using first and second images to facilitate computing a corresponding depth/disparity image can be facilitated by providing (101) the first and second images and then computing (103) a disparity value of each pixel in the second image by, at least in part, determining a likelihood of occlusion for at least some pixels comprising the second image using, at least in part, predicted occlusion information as corresponds to information contained in the first image. By one approach, this predicted occlusion information can be provided, at least in part, by processing (102) the first image, at least in part, by determining occlusion value information for at least some pixels as comprise the first image and then using the occlusion value information for the first image to determine a corresponding disparity information map for the first image.

Abstract: A device and method of modifying an image containing a foreground and an original or substitute background are disclosed. Boundary pixels contain only the original background or the original background and the foreground. The original background is replaced by a predetermined or random color or by a color corresponding to the substitute background. The entire boundary pixel can be replaced by the replacement color. Alternatively, the ratio of the foreground to the original background can be estimated from the neighboring pixels and only the original background replaced by the replacement color. Once some or all of the boundary pixels are replaced, the image can be transmitted or otherwise transferred to other devices or viewers. Some or all of the images in a video can be modified.