Abstract: An object recognition system and process that identifies people and objects depicted in an image of a scene. In general, this system and process entails first creating model histograms of the people and objects that it is desired to identify in the image. Then, the image is segmented to extract regions which likely correspond to the people and objects being identified. A histogram is computed for each of the extracted regions, and the degree of similarity between each extracted region histogram and each of the model histograms is assessed. The extracted regions having a histogram that exhibits a degree of similarity to one of the model histograms which exceeds a prescribed threshold is designated as corresponding to the person or object associated with that model histogram.

Abstract: A method, schema and computer system for a location service provides data structures that are extensible by allowing different types of location awareness applications to retrieve and deposit location data. The method includes providing an abstract location object arranged in a schema to enable servicing a received request for a logical location of an electronic endpoint; and arranging one or more location elements within the context of the abstract location object. The location elements are configured to act as peers relative to the logical location and each location element is configured to act as a proxy relative to any other location element within the abstract location object.

Abstract: A process for measuring the location of people and objects carrying radio frequency (RF) transmitters (TXs) that transmit messages to a plurality of RF receivers (RXs) located in a space. Each RX is in communication with a computer of a computer network and forwards data received from the TXs to the network via its associated computer, along with a value indicating the signal strength of the received TX transmission. The signal strengths attributable to the same transmission are used to form a locating signal strength vector which is then compared to exemplary vectors generated from signal strength readings gathered in a calibration procedure from a set of representative locations in the space. In comparing the locating vector to the exemplary vectors, constraints are enforced on movements between locations (e.g., cannot pass through walls) and to probabilistically enforce expectations on transitions between locations.

Abstract: An object recognition system and process that identifies people and objects depicted in an image of a scene. In general, this system and process entails first creating model histograms of the people and objects that it is desired to identify in the image. Then, the image is segmented to extract regions which likely correspond to the people and objects being identified. A histogram is computed for each of the extracted regions, and the degree of similarity between each extracted region histogram and each of the model histograms is assessed. The extracted regions having a histogram that exhibits a degree of similarity to one of the model histograms which exceeds a prescribed threshold is designated as corresponding to the person or object associated with that model histogram.

Abstract: A method and a system for measuring a relative position and orientation of range cameras using a movement of an object within a scene. In general, the present invention determines the relative pose between two cameras by measuring a path the movement of the object makes within a scene and calculating transformation parameters based on these measurements. These transformation parameters are used to determine the relative position of each camera with respect to a base camera. In a preferred embodiment, the present invention also includes other novel features such as a data synchronization feature that uses a time offset between cameras to obtain the transformation parameters. In addition, the present invention includes a technique that improves the robustness and accuracy of solving for the transformation parameters and an interpolation process that interpolates between sampled points if there is no data at a particular instant in time.

Abstract: The present invention leverages changes in the sensed strength of radio signals at different locations to determine a device's location. In one instance of the invention, inference procedures are used to process ambient commercial radio signals, to estimate a location or a probability distribution over the locations of a device. In an instance of the invention, learning and inference methods are applied to rank vector of signal strength vectors. Moving to such rank orderings leads to methods that bypass consideration of absolute signal strengths in location calculations. The invention facilitates approximations for locating a device by providing a method that does not require a substantial number of available ambient signal strengths while still providing useful location inferences in determining locations.

Abstract: The present invention leverages changes in the sensed strength of radio signals at different locations to determine a device's location. In one instance of the present invention, inference procedures are used to process ambient commercial radio signals, to estimate a location or a probability distribution over the locations of a device. In another instance of the present invention, a system utilizes learning and inference methods that are applied to rank vector of signal strength vectors. Moving to such rank orderings leads to systems that bypass consideration of absolute signal strengths in location calculations. The present invention facilitates approximations for locating a device by providing a system that does not require a substantial number of available ambient signal strengths while still providing useful location inferences in determining locations.

Abstract: The present invention employs approximate device locations determined from changes in the sensed strength of radio signals at different locations. In one instance of the invention, the approximate device locations are based on inference procedures that are used to process ambient commercial radio signals, to estimate a location or a probability distribution over the locations of a device. In another instance of the invention, approximate device locations derived from learning and inference methods that are applied to rank vector of signal strength vectors are utilized. Moving to such rank orderings leads to methods that bypass consideration of absolute signal strengths in location calculations. The invention utilizes approximations for a device location that is based on a method that does not require a substantial number of available ambient signal strengths while still providing useful location inferences in determining locations.

Abstract: A process for measuring the location of people and objects carrying radio frequency (RF) transmitters (TXs) that transmit messages to a plurality of RF receivers (RXs) located in a space. Each RX is in communication with a computer of a computer network and forwards data received from the TXs to the network via its associated computer, along with a value indicating the signal strength of the received TX transmission. The signal strengths attributable to the same transmission are used to form a locating signal strength vector which is then compared to exemplary vectors generated from signal strength readings gathered in a calibration procedure from a set of representative locations in the space. In comparing the locating vector to the exemplary vectors, constraints are enforced on movements between locations (e.g., cannot pass through walls) and to probabilistically enforce expectations on transitions between locations.

Abstract: A system and process for recognizing an object in an input image involving first generating training images depicting the object. A set of prototype edge features is created that collectively represent the edge pixel patterns encountered within a sub-window centered on each pixel depicting an edge of the object in the training images. Next, a Hough kernel is defined for each prototype edge feature in the form of a set of offset vectors representing the distance and direction, from each edge pixel having an associated sub-window exhibiting an edge pixel pattern best represented by the prototype edge feature, to a prescribed reference point on a surface of the object. The offset vectors are represented as originating at a central point of the kernel. For each edge pixel in the input image, the prototype edge feature which best represents the edge pixel pattern exhibited within the sub-window centered on the edge pixel is identified.

Abstract: A system and process for recognizing an object in an input image involving first generating training images depicting the object. A set of prototype edge features is created that collectively represent the edge pixel patterns encountered within a sub-window centered on each pixel depicting an edge of the object in the training images. Next, a Hough kernel is defined for each prototype edge feature in the form of a set of offset vectors representing the distance and direction, from each edge pixel having an associated sub-window exhibiting an edge pixel pattern best represented by the prototype edge feature, to a prescribed reference point on a surface of the object. The offset vectors are represented as originating at a central point of the kernel. For each edge pixel in the input image, the prototype edge feature which best represents the edge pixel pattern exhibited within the sub-window centered on the edge pixel is identified.

Abstract: A system and process that computes the probability of histogram matching false alarms for different settings of a histogram matching algorithm's parameters is presented. This allows the parameters to be adjusted to produce the optimum object finding capability with the lowest possible false alarm rate. Generally, a prototype histogram is generated from a model of an item being sought in an environment. The set of all possible test histograms that can be formed from the environment given a prescribed number of bins and a maximum count for the bins is then determined. Once this is accomplished a subset of test histograms from the set of all possible test histograms which will cause a false alarm is found. Then the probability of occurrence of each individual test histogram that will cause a false alarm in the subset is determined and summed to establish the overall false alarm probability.

Abstract: A process for measuring the location of people and objects carrying radio frequency (RF) transmitters (TXs) that transmit messages to a plurality of RF receivers (RXs) located in a space. Each RX is in communication with a computer of a computer network and forwards data received from the TXs to the network via its associated computer, along with a value indicating the signal strength of the received TX transmission. The signal strengths attributable to the same transmission are used to form a locating signal strength vector which is then compared to exemplary vectors generated from signal strength readings gathered in a calibration procedure from a set of representative locations in the space. In comparing the locating vector to the exemplary vectors, constraints are enforced on movements between locations (e.g., cannot pass through walls) and to probabilistically enforce expectations on transitions between locations.

Abstract: An object recognition system and process that identifies people and objects depicted in an image of a scene. In general, this system and process entails first creating model histograms of the people and objects that it is desired to identify in the image. Then, the image is segmented to extract regions which likely correspond to the people and objects being identified. A histogram is computed for each of the extracted regions, and the degree of similarity between each extracted region histogram and each of the model histograms is assessed. The extracted regions having a histogram that exhibits a degree of similarity to one of the model histograms which exceeds a prescribed threshold is designated as corresponding to the person or object associated with that model histogram.

Abstract: A process for determining the location of entities carrying transmitters (TXs) that transmit TX ID messages to at least one receiver (RX) connected to a computer, in a network of computers. Each RX sends data messages to its associated computer which includes an identifier identifying the TX transmitting the TX ID message, the signal strength of the received TX ID message, and a RX identifier. Each RX computer generates badge hit messages from each data message, and provides them to a centralized computer of the network. The centralized computer generates a badge hit table having a separate entry for each badge hit message, which has fields including the information received in the badge hit message and a hit time indicating the time the entry was added to the table. The badge hit table data is used to determining the location of the TXs, and so the entities.

Abstract: A system and process for making live digital video feeds simultaneously available to multiple applications running on a computing device. This involves using a live image server that maintains a camera directory file containing the names of available feeds, and the name and location of an image file for each feed. The image file includes image data corresponding to a current frame of the feed and an index indicative of a frame sequence number. Each time a new frame becomes available, image data representing the frame is put into its associated image file, in place of previously stored image data. A client application chooses a feed from the camera directory file, then accesses the associated image file and makes a copy of the image data. When the client wants a new frame, it determines if the image file index has changed. If so, the client copies the new image data.

Abstract: An object recognition system and process that identifies people and objects depicted in an image of a scene. In general, this system and process entails first creating model histograms of the people and objects that it is desired to identify in the image. Then, the image is segmented to extract regions which likely correspond to the people and objects being identified. A histogram is computed for each of the extracted regions, and the degree of similarity between each extracted region histogram and each of the model histograms is assessed. The extracted regions having a histogram that exhibits a degree of similarity to one of the model histograms which exceeds a prescribed threshold is designated as corresponding to the person or object associated with that model histogram.

Abstract: A system and method for interactively modeling a room using camera images. Images of the room are input. The floor is marked with calibration markers placed at known (x,y) locations. The user runs a program that lets him “click” on these points to establish a correspondence between their absolute (x,y) coordinates on the floor and their image coordinates in each view. These correspondences are used to compute a “homography” for each image which in turn is used to compute a warped version of each image, showing what the room would look like if viewed from above. These warped images are each rendered with respect to the same origin and are used as guides in a drawing program to drag drawings of objects to place them on top of the corresponding objects in the warped images. The warped images are then deleted, leaving behind the room's objects in their correct locations.

Abstract: This invention is directed toward an object recognition system and process that identifies the location of a modeled object in a search image. This involves first capturing model images of the object whose location is to be identified in the search image. A co-occurrence histogram (CH) is then computed for each model images. A model image CH is computed by generating counts of every pair of pixels whose pixels exhibit colors that fall within the same combination of a series of pixel color ranges and which are separated by a distance falling within the same one of a series of distance ranges. Next, a series of search windows, of a prescribed size, are generated from overlapping portions of the search image. A CH is also computed for each of these search windows using the pixel color and distance ranges established for the model image CHs. A comparison between each model image CH and each search window CH is conducted to assess their similarity.

Abstract: This invention is directed toward an object recognition system and process that identifies the location of a modeled object in a search image. This involves first capturing model images of the object whose location is to be identified in the search image. A co-occurrence histogram (CH) is then computed for each model images. A model image CH is computed by generating counts of every pair of pixels whose pixels exhibit colors that fall within the same combination of a series of pixel color ranges and which are separated by a distance falling within the same one of a series of distance ranges. Next, a series of search windows, of a prescribed size, are generated from overlapping portions of the search image. A CH is also computed for each of these search windows using the pixel color and distance ranges established for the model image CHs. A comparison between each model image CH and each search window CH is conducted to assess their similarity.