Abstract: Cameras and image processing techniques are applied to the control of HVAC systems. Occupancy is detected using head-counting or motion detection. Activities are recognized in images and image sequences by machine-recognition techniques. The nature of activities, the intensity of activities, the number of occupants and their activities, etc. are all inferred from images and image sequences and used to predict current loads and/or required control signals for regulating an HVAC system.

Abstract: A method and apparatus are disclosed for recommending items of interest to a user, such as television program recommendations, before a viewing history or purchase history of the user is available. A third party viewing or purchase history is processed to generate stereotype profiles that reflect the typical patterns of items selected by representative viewers. A user can select the most relevant stereotype(s) from the generated stereotype profiles and thereby initialize his or her profile with the items that are closest to his or her own interests. A clustering routine partitions the third party viewing or purchase history (the data set) into clusters using a k-means clustering algorithm, such that points (e.g., television programs) in one cluster are closer to the mean of that cluster than any other cluster.

Abstract: System and method for displaying a driving scene to a driver of an automobile. The system comprises at least one camera having a field of view and facing in the forward direction of the automobile. The camera captures images of the driving scene, the images comprised of pixels of the field of view in front of the automobile. A control unit receives the images from the camera and applies a salt and pepper noise filtering to the pixels comprising the received images. The filtering improves the quality of the image of the driving scene received from the camera when degraded by a weather condition. A display receives the images from the control unit after application of the filtering operation and displays the images of the driving scene to the driver.

Abstract: An appliance controller is disclosed that monitors the operation of an appliance and automatically adjusts the appliance in response to predefined audio activity. The disclosed appliance controller includes one or more microphones to capture audio information associated with an appliance. A number of exemplary appliance adjustment rules define various audio features that suggest when an appliance setting should be adjusted. Each rule contains one or more audio features and a corresponding appliance adjustment that should be performed when the rule is satisfied. The appliance adjustment rules can be predefined or dynamically determined from observed user behavior.

Abstract: Generally, techniques for analyzing foreground-segmented images are disclosed. The techniques allow clusters to be determined from the foreground-segmented images. New clusters may be added, old clusters removed, and current clusters tracked. A probabilistic framework is used for the analysis of the present invention. A method is disclosed that estimates cluster parameters for one or more clusters determined from an image comprising segmented areas, and evaluates the cluster or clusters in order to determine whether to modify the cluster or clusters. These steps are generally performed until one or more convergence criteria are met. Additionally, clusters can be added, removed, or split during this process. In another aspect of the invention, clusters are tracked during a series of images, and predictions of cluster movements are made.

Abstract: A method and apparatus are disclosed for recommending items of interest to a user, such as television program recommendations, before a viewing history or purchase history of the user is available. A third party viewing or purchase history is processed to generate stereotype profiles that reflect the typical patterns of items selected by representative viewers. A user can select the most relevant stereotype(s) from the generated stereotype profiles and thereby initialize his or her profile with the items that are closest to his or her own interests. A clustering routine partitions the third party viewing or purchase history (the data set) into clusters using a k-means clustering algorithm, such that points (e.g., television programs) in one cluster are closer to the mean of that cluster than any other cluster. A mean computation routine computes the symbolic mean of a cluster.

Abstract: A method and apparatus are disclosed for recommending items of interest to a user, such as television program recommendations, before a viewing history or purchase history of the user is available. A third party viewing or purchase history is processed to generate stereotype profiles that reflect the typical patterns of items selected by representative viewers. A user can select the most relevant stereotype(s) from the generated stereotype profiles and thereby initialize his or her profile with the items that are closest to his or her own interests. A clustering routine is disclosed to partition the third party viewing or purchase history (the data set) into clusters, such that points (e.g., television programs) in one cluster are closer to the mean of that cluster than any other cluster. A mean computation routine is also disclosed to compute the symbolic mean of a cluster.

Abstract: A method and apparatus are disclosed for monitoring a location using vision-based technologies and to automatically trigger the activation of a child safety feature or an alarm when a child is detected. One or more image capture devices are focused on a given location. The captured images are processed to identify the presence of a child and to initiate an appropriate response, such as sending assistance, activating a child safety feature or triggering an alarm. A number of rules can be utilized to define various child safety events. Each rule contains one or more conditions that must be satisfied in order for the rule to be triggered, and, optionally, a corresponding action-item that should be performed when the rule is satisfied, such as sending assistance, activating a child safety feature or triggering an alarm.

Abstract: Generally, an Eigen network and system using same are disclosed that use Principal Component Analysis (PCA) in a middle (or “hidden”) layer of a neural network. The PCA essentially takes the place of a Radial Basis Function hidden layer. A classifier comprises inputs that are routed to a PCA device. The PCA device performs PCA on the inputs and produces outputs (entitled “PCA outputs” for clarity). The PCA outputs are connected to output nodes. Generally, each output is connected to each output node. Each connection is multiplied by a weight, and each output node uses weighted PCA outputs to produce an output (entitled a “node output” for clarity). These node outputs are then generally compared in order to assign a class to the input. A system uses the PCA classifier to classify input patterns. In a third aspect of the invention, a PCA classifier is trained in order to determine weights for each of the connections that are connected to the output nodes.

Abstract: A method and apparatus are disclosed for recommending items of interest to a user based on recommendations made to one or more third parties. The recommendation scores generated by a primary recommender are influenced by recommendations generated for one or more third parties, such as a friend, colleague or trendsetter. The disclosed recommender corroborates with other recommenders when recommending items of interest and adjusts a conventional recommender score based on third party recommendations. The third party recommendations may be a top-N list of recommended items for a given third party, and may optionally include a recommendation score and an indication of whether or not the third party actually selected the recommended item. A recommender evaluates the viewing or purchase habits of a user and communicates with one or more other recommenders to determine the items that are being recommended by such other recommenders.

Abstract: A method and apparatus are disclosed for recommending items of interest to a user, such as television program recommendations, based on the viewing or purchase history of a selected third party. A viewing history of a selected third party is partitioned into a set of similar clusters. A given cluster corresponds to a segment of television programs exhibiting a specific pattern. A user can select one or more clusters from the clustered third party viewing history to supplement or replace corresponding portions (clusters) of the user's own viewing history to produce a modified viewing history. The modified viewing history is processed to generate a user profile that characterizes the viewing preferences of the user, as well as the selected viewing preferences of the third party. Program recommendations are generated using the modified user profile.

Abstract: The present invention relates to a control system and method for automatically adjusting the viewing condition to enhance the experiences associated with watching a television program. The system includes a unit for deriving a classification of each video signal received therein, a unit for observing a viewer's facial expression, and a unit for processing the output of the observed facial expression to determine whether the observed facial expression is associated with predefined facial expressions stored in a recognition module. Based on the classification and/or if there is a match between the observed facial expression with the at least one of the predefined facial expressions, the electrical power supplied to any one of electrical devices provided in the room is varied according to predetermined criteria.

Abstract: This invention provides an alarm system and method for adjusting the wake-up signals. The system includes a means for tracking the behavior of a person in a predetermined area under surveillance after the activation of an alarm clock, and a means for determining whether the person is motionless for a predetermined time period. Upon recognition that the observed behavior indicates the person is still sleeping, the wake-up signals are gradually increased. At the same time, the electrical power supplied to a plurality of electronic devices may be increased to assist the person to wake up.

Abstract: A system and method for classifying facial images from a temporal sequence of images, comprises the steps of: training a classifier device for recognizing facial images, the classifier device being trained with input data associated with a full facial image; obtaining a plurality of probe images of the temporal sequence of images; aligning each of the probe images with respect to each other; combining the images to form a higher resolution image; and, classifying said higher resolution image according to a classification method performed by the trained classifier device.

Abstract: Cameras and image processing techniques are applied to the problem of spill detection. Fixed background data may be removed using a technique based on image subtraction. The remainder of one or more images may be segmented. The segments may then be tested to identify and remove from further analysis segments corresponding to non-spill foreground objects, for example by forming luminance and/or chrominance profiles to detect the presence of a high rate of specular albido that would be associated with a wet spill. Further testing may be done, for example by determining if a segment corresponds to a planar object lying in the plane of the floor using planar projection transforms and multiple cameras. If a segment is likely to indicate a spill, a notification may be generated.

Abstract: A data-class recommender, such an electronic program guide that recommends television programs, allows users to modify their implicit profiles using the profiles of other users. For example, if a user likes the programming choices made by a friend's profile, the user can have his/her profile modified by adding parts of the friend's profile to his own, either replacing parts or forming a union of the descriptors that indicated favored classes of data. According to an embodiment, features may be labeled to allow the modifying user to select the specific parts of the friend's profile to use in making the modifications. The labeling may be done based on feature-value scores or categories for which there is a high frequency of cross-correlation with other categories in a description that defines preferred subject matter, such as a specialized description of a version space.

Abstract: A program commercial based recommendation system employing a program commercial detection module, a facial estimation module, and a program recommendation module for implementing method for developing a viewing history of a viewer is disclosed. The program commercial detection module detects commercials within a transmission signal. In response to a detection of a program commercial, the facial estimation module generates a facial estimation of a viewer to thereby determine if the viewer is watching or not watching the program commercial. In response to a generation of the facial estimation, the program recommendation module stores the program commercial within a viewing history database. The stored commercial either has a positive rating when the facial estimation indicates the viewer is watching the program commercial or a negative rating when the facial estimation indicates the viewer is not watching the program commercial.

Abstract: A data-class recommender, such an electronic program guide that recommends television programs, avoids users getting trapped in a rut when the users select the same programming material over and over again. In an embodiment, the recommender may be programmed automatically to leverage the profile of another user to broaden the user's profile. For example, the recommender may use the target descriptions of other users in a same household of the user as a guide for broadening the user's profile. Alternatively, the household profile may be used as a filter for source material for soliciting feedback from the user. In this way, rather than simply broadening the user's range arbitrarily, guidance from other profiles, related in some way to the user, is obtained and leveraged. Note that the “relationship” can include friends, published stereotypes representing interests of the user, and others.

Abstract: A multiple viewing program recommendation system employing a commercial detection module and a multiple viewing module is disclosed. In response to a generation of viewing recommendations of two or more program during the same time slot, the multiple viewing module controls a display of a recommended program having the highest viewing priority on a television screen while the recommended program is being aired on one of the television channels until the commercial detection module detects a commercial being aired on the television channel. In response to the detection of the commercial by the commercial detection module (37), the multiple viewing module controls a display of an additional recommended program having the next highest viewing priority on the television screen while the additional recommended program is being aired on another one of the television channels.

Abstract: A system and method for classifying facial images from a partial view of a facial image, the method comprising the steps of: training a classifier device for recognizing facial images, the classifier device being trained with input data associated with a facial image of a subject; detecting a partial view of a subject's facial image; warping the partial view of the subject's facial image onto a frontal image to obtain a warped image of the subject; and, classifying the warped image according to a classification method performed by the trained classifier device.