Method and Apparatus to Infer Object and Agent Properties, Activity Capacities, Behaviors, and Intents from Contact and Pressure Images

摘要

An apparatus for determining a non-apparent attribute of an object having a sensor portion with which the object makes contact and to which the object applies pressure. The apparatus has a computer in communication with the sensor portion that receives signals from the sensor portion corresponding to the contact and pressure applied to the sensor portion, and determines from the signals the non-apparent attribute. The apparatus has an output in communication with the computer that identifies the non-apparent attribute determined by the computer. A method for determining a non-apparent attribute of an object.

主权项

1. A computer-implemented method to learn classifications and properties of one or more objects by processing a sequence of surface contact and/or pressure measurements captured by a plurality of local contact or pressure sensing system at a time and over time, the method comprising the steps of:
receiving a frame or frames of the sequence which includes data for a plurality of contact or pressure measurements included in the frame; identifying one or more collections of contact and/or pressure measurements in the frame, where each collection represents an object on the surface; generating models to extract one or more features from the contact and/or pressure measurement collections that are associated with each identified object; extracting a plurality of features from the collections of contact and/or pressure measurements; classifying each of the collections of contact and/or pressure measurements using a trained or untrained classifier; supplying the extracted features and/or the object classifications from one or more frames to a machine learning engine; and using the machine learning engine to generate values for one or more properties of one or more objects and/or generate semantic representations of the behavior of one or more objects over a plurality of frames, where the machine learning engine is configured to learn properties and behavior patterns observed in the contact and/or surface pressure measurements over the plurality of frames and to identify patterns of behavior by the classified objects.