BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Analysis of Spatio-temporal Representations for Robust Footstep Re
 cognition with Deep Residual Neural Networks - Omar Costilla - Reyes (Univ
 ersity of Manchester)
DTSTART:20180125T150000Z
DTEND:20180125T160000Z
UID:TALK83441@talks.cam.ac.uk
CONTACT:Liang Wang
DESCRIPTION:Human footsteps can provide a unique behavioural pattern for r
 obust biometric systems. We propose spatio-temporal footstep representatio
 ns from floor-only sensor data in advanced computational models for automa
 tic biometric verification. Our models deliver an artificial intelligence 
 capable of effectively differentiating the fine-grained variability of foo
 tsteps between legitimate users (clients) and impostor users of the biomet
 ric system. The methodology is validated in the largest footstep database\
 , containing nearly 20\,000 footstep signals from more than 120 users. The
  database contains a large cohort of impostors and a small set of clients 
 to verify the reliability of biometric systems. We provide experimental re
 sults in 3 critical data-driven security scenarios\, according to the amou
 nt of footstep data available for model training: at airports security che
 ckpoints (smallest training set)\, workspace environments (medium\ntrainin
 g set) and home environments (largest training set). With this methodology
 \, we report state-of-the-art footstep recognition rates.\n\nBio:\nOmar Co
 stilla-Reyes Received the M.Sc. degree in Electrical Engineering from the 
 University of North Texas\, Texas\, U.S.A. in 2014. During his master stud
 ies\, he was a research assistant in projects with funding from the Nation
 al Science Foundation (NSF) and National Aeronautics and Space Administrat
 ion (NASA). His M.Sc. dissertation was on dynamic indoor positioning syste
 ms using wireless sensor networks. He is currently a research associate an
 d PhD candidate in Electrical and Electronics Engineering at the Universit
 y of Manchester\, U.K. He has published papers on applications of machine 
 learning for gait analysis in security and healthcare. His research intere
 st lies in applications of machine learning using sensor systems for secur
 ity and healthcare. He received the Best Student Paper Award in Optical Se
 nsing applications at the 2015 IEEE Sensors Conference. His Journal paper 
 entitled "Temporal Pattern Recognition in Gait Activities Recorded With a 
 Footprint Imaging Sensor System" was one of the 25 most downloaded IEEE Se
 nsors Journal papers from January to June 2017. He has won scholarships an
 d awards for academic achievement including an academic scholarship for hi
 s master’s and doctorate studies from the Mexican Science council (CONAC
 yT).
LOCATION:FW26\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR