Technion Research & Development Foundation Limited

Eth Zurich

Liberovision Ag

Tiberiu Popa

Markus H Gross

Craig Gotsman

Jean-Charles Bazin

Remo Ziegler

Richard Keiser

Marcel Germann

Claudia Kuster

A method for image processing in video conferencing, for correcting the gaze of an interlocutor in an image or a sequence of images captured by at least one real camera, comprises the steps of

Method and system for image processing in video conferencing

Estimating a pose of an articulatedD object model () by a computer is done by &#x2022;obtaining a sequence of source images () and therefrom corresponding source image segments () with objects (14) separated from the image background; &#x2022;matching such a sequence () with sequences () of reference silhouettes ('), determining one or more selected sequences of reference silhouettes (&#x2032;) forming a best match; &#x2022;for each of these selected sequences of reference silhouettes (&#x2032;), retrieving a reference pose that is associated with one of the reference silhouettes (&#x2032;); and &#x2022;computing an estimate of the pose of the articulated object model () from the retrieved reference pose or poses. The result of these steps is an initial pose estimate, which then can be used in further steps, for example, for maintaining local consistency between pose estimates from consecutive frames, and global consistency over a longer sequence of frames.

Silhouette-based pose estimation

Growing community of inventors

Zurich, Switzerland

Tiberiu Popa