Osram Gmbh

Fondazione Istituto Italiano Di Tecnologia

UniversitÀ Degli Studi Di Verona

Francesco Setti

Michael Eschey

Herbert Kaestle

Fabio Galasso

Alessio Del Bue

Irtiza Hasan

Marco Cristani

Theodore Tsesmelis

An image signal indicative of the head pose of a person in a scene, as provided by a camera may be processed by estimating, as a function of the image signal received by the image sensor, the head pose of the person. A view frustum of the person may be generated from the estimated head pose. A visual frustum of attention or VFOA of the person may be generated from the view frustum. From the visual frustum, a predicted movement trajectory of the person or an estimated scene attention map as a function of the visual frustum of attention, preferably by identifying in the attention map at least one object to be controlled may be generated.

Method of view frustum detection and a corresponding system and product

Growing community of inventors

Rovereto, Italy

Francesco Setti