Abstract: A pseudo-inverse derived model-point-to-image-point transformation hypothesis generator as could be used in a computer vision system. Preferred embodiments include generation of n-point transformations directly from n point pair sets and generation of n-point transformations from sets of n point pairs defined by a preliminary three-point transformation plus sets of (n−3) points having minimal error with respect to the three-point trnasformation. Systems include salient corner extraction and hypothesis transformation verification by further sets of point pairs.

Abstract: A method for deriving a two-dimensional first-range data model at a distance, d, from a two-dimensional second-range data model at a distance, d.sub.0. The method (10) comprises the steps of first smoothing the second-range data model, I, by masking the second-range data model by a masking matrix (step 24) and next undersampling the smoothed second-range data model (step 26) to yield a first-range data model. Both steps for smoothing (step 24) and undersampling (step 26) use operators depending on the ratio of the first-range distance, d, to the second-range distance, d.sub.0. One aspect of the invention includes a distance relating system (80) that performs the smoothing step (step 24) and undersampling step (step 26) to generate first-range data models from second-range data models in an intelligent sensor system.

Abstract: The present invention is a video answer machine which includes a video data recorder (26) for accepting and recording an outgoing video message, an audio data recorder (22) for accepting and recording an outgoing audio message, and data storage (19) for storing said outgoing video message and said outgoing audio message. Also included in he video answering machine is a control module (12) for accepting an incoming call from a caller, for determining a device type associated with the incoming call, for responding to the incoming call using either the outgoing video message or the outgoing audio message in accordance with the associated device type and for accepting either an audio response or a video response from the caller in accordance with said associated device type.

Abstract: A method for deriving a two-dimensional first-range data model at a distance, d, from a two-dimensional second-range data model at a distance, d.sub.0. The method (10) comprises the steps of first smoothing the second-range data model, I, by masking the second-range data model by a masking matrix (step 24) and next undersampling the smoothed second-range data model (step 26) to yield a first-range data model. Both steps for smoothing (step 24) and undersampling (step 26) use operators depending on the ratio of the first-range distance, d, to the second-range distance, d.sub.0. One aspect of the invention includes a distance relating system (80) that performs the smoothing step (step 24) and undersampling step (step 26) to generate first-range data models from second-range data models in an intelligent sensor system.

Abstract: An image matching method that uses corners of objects, rather than their edges, to determine if the objects match. Corner locations are obtained by finding changes of curvature in the edge boundaries of the image. The corners of a reference image are used to create a distance array, which is used as a look-up table to determine how far each corner of a test array is from a corresponding corner of the reference array. A close corner-to-corner relationship indicates a good match.

Abstract: An apparatus and method for improved segmentation and object recognition which describes objects and background appropriately based on calculations of desired local surface geometries such as local surface orientation, local surface curvature, surface extent and occluding boundaries, from sensed relative range images. Relative range data is used rather than converting such data to absolute range and then performing the desired geometric calculations.