ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Exploring Transfer Learning Approaches for Head Pose Classification from Multi-view Surveillance Images

Rajagopal, Anoop Kolar and Subramanian, Ramanathan and Ricci, Elisa and Vieriu, Radu L and Lanz, Oswald and Kalpathi, Ramakrishnan R and Sebe, Nicu (2014) Exploring Transfer Learning Approaches for Head Pose Classification from Multi-view Surveillance Images. In: INTERNATIONAL JOURNAL OF COMPUTER VISION, 109 (1-2, S). pp. 146-167.

Full text not available from this repository. (Request a copy)
Official URL: http://dx.doi.org/10.1007/s11263-013-0692-2


Head pose classification from surveillance images acquired with distant, large field-of-view cameras is difficult as faces are captured at low-resolution and have a blurred appearance. Domain adaptation approaches are useful for transferring knowledge from the training (source) to the test (target) data when they have different attributes, minimizing target data labeling efforts in the process. This paper examines the use of transfer learning for efficient multi-view head pose classification with minimal target training data under three challenging situations: (i) where the range of head poses in the source and target images is different, (ii) where source images capture a stationary person while target images capture a moving person whose facial appearance varies under motion due to changing perspective, scale and (iii) a combination of (i) and (ii). On the whole, the presented methods represent novel transfer learning solutions employed in the context of multi-view head pose classification. We demonstrate that the proposed solutions considerably outperform the state-of-the-art through extensive experimental validation. Finally, the DPOSE dataset compiled for benchmarking head pose classification performance with moving persons, and to aid behavioral understanding applications is presented in this work.

Item Type: Journal Article
Publisher: SPRINGER
Additional Information: Copyright for this article belongs to the SPRINGER, VAN GODEWIJCKSTRAAT 30, 3311 GZ DORDRECHT, NETHERLANDS
Keywords: Transfer learning; Multi-view head pose classification; Varying acquisition conditions; Moving persons
Department/Centre: Division of Electrical Sciences > Electrical Engineering
Date Deposited: 12 Jul 2014 15:07
Last Modified: 12 Jul 2014 15:07
URI: http://eprints.iisc.ac.in/id/eprint/49404

Actions (login required)

View Item View Item