On a Large Sequence-Based Human Gait Database [chapter]

J. D. Shutler, M. G. Grant, M. S. Nixon, J. N. Carter
2004 Applications and Science in Soft Computing  
Biometrics today include recognition by characteristic and by behaviour. Of these, face recognition is the most established with databases having evolved from small single shot single view databases, through multi-shot multi-view and on to current video-sequence databases. Results and potential of a new biometric are revealed primarily by the database on which new techniques are evaluated. Clearly, to ascertain the potential of gait as a biometric, a sequencebased database consisting of many
more » ... nsisting of many subjects with multiple samples is needed. A large database enables the study of inter-subject variation. Further, issues concerning scene noise (or non-ideal conditions) need to be studied, ideally with a link between ground truth and application based analysis. Thus, we have designed and built a large human gait database, providing a large multi-purpose dataset enabling the investigation of gait as a biometric. In addition, it is also a useful database for many still and sequence based vision applications.
doi:10.1007/978-3-540-45240-9_46 fatcat:evfq4ngf3fhofhpsy3p7jlqrym