3D hand tracking for human computer interaction.pdfVIP

Image and Vision Computing 30 2012 236–250 Contents lists available at SciVerse ScienceDirect Image and Vision Computing journal home ☆ 3D hand tracking for human computer interaction Victor Adrian Prisacariu ?, Ian Reid Department of Engineering Science, University of Oxford, United Kingdom a r t i c l e i n f o a b s t r a c t Article history: We propose a real-time model-based 3D hand tracker that combines image regions and the signal from an Received 8 July 2011 off-the-shelf 3-axis accelerometer placed on the users hand. The visual regions allow the tracker to cope Received in revised form 9 December 2011 with occlusions, motion blur and background clutter, while the latter aids with the inherent silhouette- Accepted 4 January 2012 pose ambiguities. The accelerometer and tracker are synchronised by casting the calibration problem as one of principal component analysis. Based on the assumption that, often, the number of possible hand con- Keywords: ?gurations is limited by the activity the hand is engaging in, we use a multiclass pose classi?er to distinguish Hand tracking Region based tracking between a number of activity dependent articulated hand con?gurations. We demonstrate the bene?ts of our 3d tracking method, both qualitatively and quantitatively, on a variety of video sequences and hand con?gurations and Real time 3d tracking show a proof-of-concept human computer interface based on our system. Accelerometer tracker integration ? 2012 Elsevier B.V. All rights reserved. Pose classi?cation Articulated hand tracking 1. Introduction embedding it inside a level set function, such as a signed distance function. Rigid pose recovery is done by maximising the posterior As computers have metamorphosed from pure computational ma- per-pixel probability of foreground and background membership as chines into domestic gaming and media hubs, at the centre of our so- a