Title: A Hierarchical Based Framework for Multimedia Forensics
| Location | Attendance |
|---|---|
| EE Department, Auditorium 10-301, SSE Building | IEEE Members: 12 |
| LUMS, DHA | Guests: 15 |
| Lahore, Punjab | Cost: false |
| Pakistan |
Map Link:
Start Time: 2011-01-06 15:00:00
End Time: 2011-01-06 16:00:00
Time Zone: Asia/Karachi
Meeting Description:
Dept. of Electrical Engineering LUMS School of Science & Engineering
A Hierarchical Based Framework for Multimedia Forensics
Bio of Speaker
Hafiz Malik is Assistant Processor in the Electrical and Computer
Engineering (ECE) Department at University of Michigan – Dearborn.
His research in multimedia forensics and security, wireless sensor
networks, steganography/steganalysis, and biometric security is
funded by the National Academies and other agencies. He has
published more than 35 papers in leading journals, conferences, and
workshops. He is on the Review Board Committee of
IEEE Technical Committee on Multimedia Communications (MMTC).
He organized Special Track on Doctoral Dissertation in Multimedia,
in the 6th
IEEE International Symposium on Multimedia (ISM) 2006. He is
serving as secretary of
IEEE SEM, Chapter III since 2009 and is also serving on
several technical program committees, including the
IEEE AVSS, ICME ICIP, MINES, ISPA, CCNC, and ICC.
Further information about his research can be found at
http://www-personal.engin.umd.umich.edu/~hafiz/
Abstract
There is rapidly growing use of digital media (images, audio, and
video recordings) as evidence in every sector of litigation,
journalism, and law enforcement applications. The integrity
authenticity of the digital media content must, however, be
ascertained given the ease with which the content can be altered
using widely available editing tools. Integrity authentication of
digital media is posing unique challenges to forensic science in
deterring the impact of digital forgery. However, available
state-of-the-art methods are limited in their scope. They are unable
to perform critical tasks in authentication such as determining if
the evidentiary recording was assembled by manipulating and splicing
original recordings captured in different environments and using
different devices.
This talk first surveys state-of-the-art in the area of multimedia
forensics. This talk will then present a hierarchical based
framework to the general problem of multimedia forensics. The focus
of this talk will be on: 1) mathematical modeling of digital audio
recording, 2) efficient separation of contributing sources from
digital recordings, 3) parameter estimation from each contributing
component, and 4) application of estimated parameters to
authenticate integrity of digital recordings, provide a high-level
description of the recording environment, and identify acquisition
device from recording. This talk will also discusses key problems
ranging from detecting forgeries in digital recordings to unique
mapping of digital recording to the acquisition device, recording
environment identification, securing speaker recognition (SR)
systems against record-and-play (RAP) attacks, and recent findings
in these areas.
Ishtiaq Bhatti +9242-3560-8312