TECHNICAL PROGRAM

WIFS'12 Schedule
The pdf version of WIFS'12 Technical Program can be downloaded here.

NOTE: This schedule may be subject to slight changes. Check this page for updates.

WIFS'12 Detailed Technical Program

NOTE: This program may be subject to slight changes. Check this page for updates.

• Sunday 2nd December 2012
  

  08:00-09:15	Registration
  09:15-12:30	Tutorial 1: A primer on content protection systems G. Doërr, D. Alain, and T. Kalker. Tutorial 2: Cloud computing for network forensics J. François.
  12:30-14:00	Lunch
  14:00-17:15	Tutorial 3: Biometric privacy and secrecy from information-theoretical perspective T. Ignatenko. Tutorial 4: Uncovering the processing history of video content A. Piva and S. Milani.

• Monday 3rd December 2012
  

  08:00-08:40	Registration
  08:40-09:00	Welcome
  09:00-10:00	Keynote 1: Adversary aware signal processing Prof. Dr. M. Barni.
  10:00-12:20	Oral Session 1: Biometrics
  10:00-10:20	Detecting Fingerprint Distortion from a Single Image Xuanbin Si, Jianjiang Feng and Jie Zhou
  10:20-10:40	Novel Fingerprint Aging Features Using Binary Pixel Sub-Tendencies: A Comparison of Contactless CLSM and CWL Sensors Ronny Merkel, Jana Dittmann and Claus Vielhauer
  10:40-11:00	Coffee break
  11:00-11:20	Facial Landmark Configuration for Improved Detection Chengwei Huang, Boris Efraty, Uday Kurkure, Xi Zhao, Manos Papadakis, Shishir Shah and Ioannis Kakadiaris
  11:20-11:40	Two-stage Appearance-based Re-Identification of Humans in Low-Resolution Videos Jürgen Metzler
  11:40-12:00	Biometric Template Protection using Turbo Codes and Modulation Constellations Emanuele Maiorana, Daniele Blasi and Patrizio Campisi
  12:00-12:20	Can a "Poor" Verification System be a "Good" Identification System? A Preliminary Study Brian DeCann and Arun Ross
  12:20-14:00	Lunch
  14:00-15:20	Poster Session 1: Security and Privacy
  14:00-15:20	Machine Learning Attacks on 65nm Arbiter PUFs: Accurate Modeling poses strict Bounds on Usability Gabriel Hospodar, Roel Maes and Ingrid Verbauwhede Towards reproducible results in authentication based on physical non-cloneable functions: the forensic authentication microstructure optical set (FAMOS) Sviatoslav Voloshynovskiy, Maurits Diephuis, Fokko Beekhof, Oleksiy Koval and Bruno Keel. Improving the DGK comparison protocol Thijs Veugen General Function Evaluation in a STPC Setting via Piecewise Linear Approximation Tommaso Pignata, Riccardo Lazzeretti and Mauro Barni A Framework for Privacy Preserving Statistical Analysis on Distributed Databases Bing-Rong Lin, Ye Wang and Shantanu Rane Modeling and Analysis of Electric Network Frequency Signal for Timestamp Verification Ravi Garg, Avinash L. Varna and Min Wu Telephone Handset Identification by Feature Selection and Sparse Representations Yannis Panagakis and Constantine Kotropoulos Privacy-Preserving Architecture for Forensic Image Recognition Andreas Peter, Thomas Hartmann, Sascha Müller and Stefan Katzenbeisser Sequence Detection of Overlapping Latent Fingerprints Using a Short-term Aging Feature Maik Schott, Ronny Merkel and Jana Dittmann
  15:20-17:40	Oral Session 2: Network Forensics
  15:20-15:40	Semantic based DNS Forensics Samuel Marchal, Jérôme François, Radu State and Thomas Engel
  15:40-16:00	Fingerprinting a flow of messages to an anonymous server Juan A. Elices and Fernando Pérez-González
  16:00-16:20	Coffee break
  16:20-16:40	Detecting anomalous network hosts by means of PCA Tomáš Pevný, Martin Rehák and Martin Grill
  16:40-17:00	Forensic Analysis of Wireless Networking Evidence of Android Smartphones Panagiotis Andriotis, George Oikonomou and Theo Tryfonas
  17:00-17:20	A Least Squares Approach to User Profiling in Pool Mix-based Anonymous Communication Systems Fernando Pérez-González and Carmela Troncoso
  17:20-17:40	Mapping Evidence Graphs to Attack Graphs Changwei Liu, Anoop Singhal and Duminda Wijesekera
  17:40-19:00	Ongoing Work Session
  17:40-19:00	Demo Session
  17:40-19:00	A Tattoo image retrieval system Daniel Manger Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB Abstract: Within the EU research project FASTID (FAST and efficient international disaster victim IDentification), a system for large-scale tattoo image retrieval has been developed. Given a database of images, the target is to determine whether a query image shows a tattoo which is depicted in one of the database images. The system is tuned to work with real images or template images of tattoos but it also works remarkably well in many other domains of forensics and security. In this interactive demonstration, the image retrieval system is presented - prepared with some 100.000 tattoo images from the web. However, the audience can add any picture they want to the database by taking an image from the web or using a camera to capture an image of an object (business card, jewellery, and of course tattoo - if available). For querying the database, images corresponding to the tattoos in the database are used. Again, using the camera, the audience may use any image to query the database resulting in a list of the most similar images. By enabling the interactivity for both the database and query images, the strengths and limits of a modern content-based image retrieval system can be explored. REWIND-Demos@WIFS2012 Paolo Bestagini, Pedro Comesaña Alfaro, Christian Dittmar, Marco Fontani, Marco Visentini Scarsella, and Stefano Tubaro REWIND project (http://www.rewindproject.eu/) Abstract: REWIND (REVerse engineering of audio-VIsual coNtent Data, www.rewindproject.eu) is a research project funded by the European Commission within the FP7 ICT FET-Open program. The project is aimed at synergistically combining principles of signal processing, machine learning and information theory to answer relevant questions on the past history of multimedia objects. The latter can be expressed in terms of the processing chains applied to the considered object, which can be very complex. The REWIND project relies on the fact that each one of the considered processing steps is bound to leave a characteristic footprint that can be detected and analyzed in a blind fashion (i.e. without having access to the content at any earlier point of the processing chain), creating the opportunity of tracing back the past history. So far, REWIND has obtained significant results that can have impact in different application areas. In particular, during the IEEE WIFS2012 Demo Session, REWIND will present a set of different tools in the area of multimedia forensics. More precisely it will be demonstrated the detection of: image tampering, multiple JPEG compression, audio plagiarism, video recapturing and video source identification. An Effective Tool for Copy-Move Forgery Detection and Localization Irene Amerini, Lamberto Ballan, Roberto Caldelli, Alberto Del Bimbo, Luca Del Tongo, Giuseppe Serra Media Integration and Communication Center, University of Florence Abstract: We demonstrate a novel image forensics tool for copy-move detection and localization. The proposed method can determine if such tampering has occurred and which image patches are involved, and it also recovers which geometric transformation was used to perform cloning. Technicolor’s ContentArmorTM Solutions Michael Arnold, Séverine Baudry, Peter G. Baum, Xiao-Ming Chen, Bertrand Chupeau, Gwenaël Doërr, Ulrich Gries, Frédéric Lefèbvre, Sylvain Lelievre, Antoine Robert, Christophe Vincent and Mario de Vito Technicolor Abstract: Through its expertise in security technologies and entertainment services, Technicolor helps studios, production companies and distributors to maximize the value of their creative assets. Thanks to Technicolor’s state-of-the-art security solutions, each link in the content value chain is protected. ContentArmor™ is Technicolor's own full-fledge content protection technology which combines password-based access control, visual overlays, forensic audio and video watermarking, and video fingerprinting. ContentArmor™ secures processes and digital file-based services; typical use cases include secure transfer, audio-dubbing, screeners, dailies and more. In this demo, the visitor will be personally walked through the whole ContentArmor™ pipeline including content packaging, secure content viewing and watermarking-based investigative services. Demonstration of the Gradiant Asymmetric Encryption System based on Handwritten Signature Enrique Argones Rúa, Francisco Javier García Salomón, Luis Pérez Freire Gradiant Abstract: In this demonstration a system for encryption and digital signature of generic digital documents is presented, where cryptographic keys are protected by a secured (protected, reliable and revokable) hand-written signature of the user. Biometric template protection of the system is based on the scheme presented in [1], with an improved feature combination scheme that increases the template protection strength, providing 99 bits of protection instead of the 73 bits offered by the scheme presented in the paper. The demonstration will show the typical user interactions with the system, which will be installed in a laptop. The device used for capturing signatures is a Wacom Intuos 4 S. Our main objective is to present the technology and to have feedback from the community. Therefore the public, after a brief presentation of the technology, will be invited to interact with the system, performing enrollments, tests (document sign, encrypt and decrypt operations) or even trying to spoof the system by imitating another's user signature. [1] E. Argones R´ua, E. Maiorana, J.L. Alba Castro, and P. Campisi, “Biometric template protection using universal background models: An application to online signature,” IEEE Trans. Inf. Forensics Security, vol. 7, no. 1, pp. 269–282, 2012. Watermarking technology for ambient detection on mobile devices Tomas Filler and Osama Alattar Digimarc Abstract: Digimarc will demo its audio watermarking technology aimed at ambient detection on mobile devices (smartphones, tablets) in real-world environments (living rooms, cars, etc). This technology showcases the results of our latest research into psychoacoustic models, reliable detection in ambient conditions, and fast detection on computationally constrained platforms. The demo will illustrate the use of audio watermarking for content identification within our Digimarc Discover platform. This platform integrates multiple content identification technologies to give mobile devices the capability to see, hear, and engage with all types of media. We will show how this platform enables users to discover more about audio, video, printed/digital images, and explicit symbology (barcodes). SIGNED: a tool for protecting the authenticity of digital and printed documents Claudio Dosio, Alberto Malvido, Kush Wadwha, Luis Pérez Freire, Giuseppe Italiano SIGNED project (www.signedfp7.eu) Abstract: SIGNED is a project financed by the European Commission for developing a paper-based document authentication system, making use of image hashing techniques and public key cryptography, which fills the gap in the digital trust process created by the need of printing documents. The final aim of the SIGNED project is to provide an extension of the digital signature technology suitable to cope with paper-based documents. The demonstration to be performed in WIFS will showcase the SIGNED prototype developed during the project, providing to the audience the opportunity to try the processes of creating a protected document and veryfying the authenticity of a printed document. Monitoring Large Scale Networks using Tree Based Multidimensional Aggregation Lautaro Dolberg, Jérôme François, Thomas Engel SnT - Interdisciplinary Centre for Security, Reliability and Trust, University of Luxembourg Abstract: In this demo, we introduce an interactive a tool for large scale network data aggregation. Our aggregation technique is a novel technique supporting multiple dimensions i.e different kinds of feature, like IP address or host locations without providing initial directives for data space partitioning for aggregation. Aggregation provides the ability of highlighting important changes in the network activity in large volume of data nomalies and attacks) which is essential in the context of network forensics. Validation was done by measuring the impact of the aggregation process trough multiple scenarios based on network related data containing multiple dimensions. During the demo multiple test covering different scenarios are conducted for illustrating multiple dimensions can be aggregated. Dimensions included in our tests are: source and destination IP addresses, services, geographical location of hosts, DNS names, etc. The tool output can be visualized as a tree shaped structure or GPS coordinates displayed using Google Maps. Scenarios can be adapted based on the interactions with the audience. As our tool will be deployed in our own laptop, the only equipment that can be helpful for presentation is a big monitor for improving the visibility.

• Tuesday 4th December 2012
  

  08:00-09:00	Registration
  09:00-10:00	Keynote 2: Playing hide and seek online: anonymous communications, traffic analysis and censorship G. Danezis
  10:00-12:20	Oral Session 3: Multimedia Forensics
  10:00-10:20	Reverse Engineering of Double JPEG Compression in the Presence of Image Resizing Tiziano Bianchi and Alessandro Piva
  10:20-10:40	Content Authentication and Identification under Informed Attacks Fokko Beekhof, Sviatoslav Voloshynovskiy and Farzad Farhadzadeh
  10:40-11:00	Coffee break
  11:00-11:20	Forensic Analysis of Ordered Data Structures On The Example of JPEG Files Thomas Gloe
  11:20-11:40	Preventing unauthorized copying of displayed information by utilizing differences in spectral sensitivity between humans and imaging devices Takayuki Yamada, Seiichi Gohshi and Isao Echizen
  11:40-12:00	Detection of video double encoding with GOP size estimation David Vázquez-Padín, Marco Fontani, Tiziano Bianchi, Pedro Comesaña, Alessandro Piva, and Mauro Barni
  12:00-12:20	IPTV streaming source classification Miguel Masciopinto, Pedro Comesaña
  12:20-14:00	Lunch
  14:00-20:30	Visit to Observatory
  20:30-	Banquet (Location)

• Wednesday 5th December 2012
  

  08:00-09:00	Registration
  09:00-10:00	Keynote 3: IEEE Biometrics Council Distinguished Lecturer Talk: Issues, Controversies and Advancements in Forensic Speaker Recognition J. L. Wayman
  10:00-12:20	Oral Session 4: Data Hiding
  10:00-10:20	Effect of Cover Quantization on Steganographic Fisher Information Jessica Fridrich
  10:20-10:40	An Efficient JPEG Steganographic Scheme Using Uniform Embedding Linjie Guo, Jiangqun Ni and Yun Qing Shi
  10:40-11:00	Coffee break
  11:00-11:20	Active content fingerprinting: a marriage of digital watermarking and content fingerprinting Sviatoslav Voloshynovskiy, Farzad Farhadzadeh, Oleksiy Koval and Taras Holotyak
  11:20-11:40	AC-3 Bit Stream Watermarking Xiao-Ming Chen, Michael Arnold, Peter Baum and Gwenaël Doërr
  11:40-12:00	Decoding Fingerprints Using the Markov Chain Monte Carlo Method Teddy Furon, Arnaud Guyader and Frédéric Cérou
  12:00-12:20	Weighted Stego-Image Steganalysis for Naive Content-Adaptive Embedding Pascal Schoettle, Stefan Korff and Rainer Boehme
  12:20-14:00	Lunch
  14:00-15:20	Poster Session 2: Media Forensics and Security
  14:00-15:20	Optimum Forensic and Counter-forensic Strategies for Source Identification with Training Data Mauro Barni and Benedetta Tondi ML Estimation of the Resampling Factor David Vázquez-Padín, Pedro Comesaña Detection and information theoretic measures for quantifying the distinguishability between multimedia operator chains Pedro Comesaña Watermarking Method with Exact Self-Propagating Restoration Capabilities Sergio Bravo-Solorio, Chang-Tsun Li and Asoke Nandi Capacity Control of Reversible Watermarking by Two-Thresholds Embedding Ion Caciula and Dinu Coltuc How to Find Relevant Training Data: A Paired Bootstrapping Approach to Blind Steganalysis Pham Hai Dang Le and Matthias Otto Franz Designing Steganographic Distortion Using Directional Filters Vojtěch Holub and Jessica Fridrich Dynamic Traitor Tracing for Arbitrary Alphabets: Divide and Conquer Thijs Laarhoven, Jan-Jaap Oosterwijk and Jeroen Doumen Low Level Crowd Analysis Using Frame-Wise Normalized Feature For People Counting Hajer Fradi and Jean-Luc Dugelay Identify computer generated characters by analysing facial expressions variation Duc-Tien Dang-Nguyen, Giulia Boato and Francesco G.B. De Natale
  15:20-17:20	Oral Session 5: Secure Communications
  15:20-15:40	Non-regenerative Two-Hop Wiretap Channels using Interference Neutralization Sabrina Gerbracht, Eduard Jorswieck, Gan Zheng and Björn Ottersten
  15:40-16:00	End-to-end secure connection in heterogeneous networks for critical scenarios Romeo Giuliano, Alessandro Neri and Damiano Valletta
  16:00-16:20	Coffee break
  16:20-16:40	Comparison of Different Attack Classes in Arbitrarily Varying Wiretap Channels Holger Boche and Rafael Felix Wyrembelski
  16:40-17:00	Optimality of Beamforming for Secrecy Capacity of MIMO Wiretap Channels Jiangyuan Li and Athina Petropulu
  17:00-17:20	Destination Assisted Cooperative Jamming for Wireless Physical Layer Security Yupeng Liu and Athina Petropulu
  17:20-17:30	Conclusions and Farewell