CAST-Workshop

Biometrie-BIOSIG 2017

Termin: 20.-22.09.2017
Dauer: 10:00-17:00
Ort: Fraunhofer IGD
Fraunhoferstr. 5
64283 Darmstadt

Programm

Wednesday, September 20
17:30
Claudia Prediger
CAST e.V.
Claudia Prediger
BIOSIG registration
19:30
-
EAB Award ceremony and BIOSIG welcome reception

Thursday, September 21


BIOSIG - MAIN CONFERENCE -

10:00
Christoph Busch
Hochschule Darmstadt
Christoph Busch
BIOSIG Conference Opening
10:10
Ciaran Carolan
eu-LISA
Ciaran Carolan
KEYNOTE
Biometrics, interoperability and large-scale IT systems: the Future of EU Smart Borders

Border management procedures worldwide are increasingly reliant on technologies and digital systems. European authorities involved in policy formulation for border control have increasingly looked to IT solutions as they seek appropriate means to deal with increasing flows of travellers at the external borders of the Schengen Area while striving to render border checks more efficient and effective. As part of the European Commission’s Smart Borders program, two new large-scale IT systems have been proposed for development. The European Entry Exit System (EES) aims to facilitate secure border passage through the use of biometrics and self-service technologies. The European Travel Information Authorisation System (ETIAS), meanwhile, will allow collection of data for risk assessment in advance of travel. The new systems complement those already in place, with interoperability between systems to improve service efficiency and the effectiveness of system usage a relatively new area of interest.

The latest developments in the expansion of IT systems at European borders will be described in the presentation. Updates will be provided on eu-LISA-led studies on system development and interoperability concepts. The information will provide the basis for a depiction of the future of IT-led border management encompassing advance passenger authorization as well as checks at borders and involving a variety of actors including border guards, customs authorities, consular authorities, travel agents and carriers and of course travellers.

10:50
Gabriel Emile Hine
Roma Tre University
Gabriel Emile Hine
Resting-state EEG
A Study on its non-Stationarity for Biometric Applications
  • EEG
  • Biometrics
  • non-Stationarity
11:20 Break
11:55
Aske Rasch Lejbølle
Aalborg University
Aske Rasch Lejbølle
Multimodal Neural Network for Overhead Person Re-identification
  • Multimodal
  • Person Re-identification
  • Convolutional Neural Networks
  • Feature Fusion
12:25
Nanang Susyanto
Universitas Gadjah Mada
Nanang Susyanto
Pool Adjacent Violators Based Biometric Rank Level Fusion
  • rank level fusion
  • pool adjacent violators
  • biometric fusion
  • biometric identification
  • multimodal
12:55 Lunch-Break
14:05
Christian Rathgeb
Hochschule Darmstadt
Christian Rathgeb
Improvement of Iris Recognition based on Iris-Code Bit-Error Pattern Analysis
  • iris recognition
  • iris-code
  • bit-error analysis
  • improved comparator
14:35
Ehsaneddin Jalilian
University of Salzburg
Ehsaneddin Jalilian
Domain Adaptation for CNN Based Iris Segmentation
  • Iris Segmentation
  • Damian Adaptation
  • CNN based Iris Segmentation
15:05
Pawel Drozdowski
Hochschule Darmstadt
Pawel Drozdowski
SIC-Gen: A Synthetic Iris-Code Generator
  • Biometrics
  • Iris Recognition
  • Iris-Code
  • Synthetisation
15:35 Break
16:10
Johannes Merkle
secunet
Johannes Merkle
xT ARP
Improving the T ented Arch Reference Point Detection Algorithm
  • Fingerprint registration
  • Reference point detection
  • Fingerprint alignment
  • Biometric template protection
16:40
Simon Kirchgasser
University of Salzburg
Simon Kirchgasser
Fingerprint T emplate Ageing vs. T emplate Changes Revisited
  • Fingerprint Recognition
  • Template Ageing
  • Quality
  • Ghost Fingerprints
17:10
Arslan Brömme
Opening Poster Session

1

Andre Ricardo Goncalves
Download Paper (only for members)

CPqD

On the generalization of fused systems in voice presentation attack detection

2

Eduardo Ribeiro
Download Paper (only for members)

University of Salzburg

Exploring Texture Transfer Learning via Convolutional Neural Networks for Iris Super Resolution

3

Patrick Schuch
Download Paper (only for members)

NTNU

Intrinsic Limitations of Fingerprint Orientation Estimation

4

Neamah Al-Naffakh
Download Paper (only for members)

Plymouth University

Unobtrusive Gait Recognition using Smartwatches

5

Attaullah Buriro
Download Paper (only for members)

University of Trento

Evaluation of Motion-based Touch-typing Biometrics in Online Financial Environments

6

Emanuela Piciucco
Download Paper (only for members)

Roma Tre University

Steady-State Visual Evoked Potentials for EEG-Based Biometric Identification

7

Fernando Alonso-Fernandez
Download Paper (only for members)

Halmstad University

Improving Very Low-Resolution Iris Identification Via Super-Resolution Reconstruction of Local Patches

8

Andreas Nautsch
Download Paper (only for members)

Hochschule Darmstadt

Deep Quality-informed Score Normalization for privacy-friendly Speaker Recognition in unconstrained Environments

9

Francisco M. Castro
Download Paper (only for members)

University of Málaga

Evaluation of CNN architectures for gait recognition based on optical flow maps

10

Chris Zeinstra
Download Paper (only for members)

University of Twente

How Random is a Classifier given its Area under Curve?

11

Stepanka Barotova
Download Paper (only for members)

Brno University of Technology

Fingerprint Damage Localizer and Detector of Skin Diseases from Fingerprint Images

12

Pankaj Wasnik
Download Paper (only for members)

NTNU

Fusing Biometric Scores using Subjective Logic for Gait Recognition on Smartphone

18:40 Social Event: Dinner with Barbeque

Friday, September 22

09:00
Yoshinori Koda
NEC Corporation
Yoshinori Koda
KEYNOTE
Fingerprinting young newborns - from sensor design to application challenges

The use of the technology for identification has recently expanded due to several innovations in the field. One such expansion is the use of biometrics to create a national ID system as a HUB for several social services. Such a system would be mainly used for history management of issuing social welfare services for citizens, especially for adults. However, additional requirements requested involve development of the ability to register children into the citizenship database. A biometric technology solution makes identification possible because it does not require any input from the child.

NEC, a biometric technology pioneer, assigned a technical project team to develop an accurate child print capturing method without the limits prescribed by the various international standards and specifications. This project is in line with the company’s corporate philosophy of providing innovative solutions for society’s most challenging problems. The purpose of this research is to explain the feasibility of NEC’s new device with biometric technology and the results from onsite research.

09:40
Vanina Camacho
Universidad de la República
Vanina Camacho
Recognizing infants and toddlers over an on-production fingerprint database
  • fingerprints
  • biometrics
  • newborns
  • infants
  • recognition
10:10
Tarang Chugh
Michigan State University
Tarang Chugh
Benchmarking Fingerprint Minutiae Extractors
  • fingerprint recognition
  • minutiae extraction
  • robustness to noise
  • interoperability
10:40 Break
11:15
Luuk Spreeuwers
University of Twente
Luuk Spreeuwers
De-duplication using automated face recognition: an accurate mathematical model and all babies are equally cute
  • de-duplication
  • face recognition
  • large database
  • binomial distribution
11:45
Seong Tae Kim
KAIST
Seong Tae Kim
Multi-scale facial scanning via spatial LSTM for latent facial feature representation
  • face recognition
  • facial feature representation
  • spatial LSTM
  • deep learning
12:15
Mehmet Ozgur Turkoglu
University of Twente
Mehmet Ozgur Turkoglu
Texture-Based Eyebrow Recognition
  • eyebrow recognition
  • texture-based
  • local descriptor
  • WLD
  • 3-patch LBP
  • 4-patch LBP
12:45 Lunch-Break
13:55
Ulrich Scherhag
Hochschule Darmstadt
Ulrich Scherhag
Biometric Systems under Morphing Attacks: Assessment of Morphing Techniques and Vulnerability Reporting
  • Biometrics
  • Morphing
  • Performance Reporting
  • Attack Detection
14:25
Sushil Bhattacharjee
Idiap Research Institute
What you can't see can help you
Extended-range imaging for 3D-mask presentation attack detection
  • Presentation Attack Detection
  • 3D-Masks
  • Extended-range imaging
  • Near Infrared (NIR)
  • Long-wave Infrared (LWIR)
  • Thermal imagery
14:55
Héctor Delgado
EURECOM
Héctor Delgado
Impact of bandwidth and channel variation on presentation attack detection for speaker verification
  • presentation attack detection
  • speaker verification
  • bandwidth and channel variation
15:25 Break
16:00
Mark Nixon
University of Southampton
Mark Nixon
KEYNOTE
Soft Biometrics: Identification by Face, Body and Clothing

Soft Biometrics are the estimation or use of personal characteristics describable by humans that can be used to aid or effect person recognition. Soft biometrics are also attributes that are typically gleaned from primary biometric data which are classifiable in pre-defined human understandable categories and can be extracted in an automated manner. These attributes can contribute to recognition or can be estimated separately and this helps to improve or refine biometric systems operation. This leads to soft biometrics which can be used to reinforce conventional biometrics and which can be estimated separately, say to increase biometrics deployment capability. Alternatively, fine grained attributes can be deployed as measures used for identification. Humans can describe the body, the clothing and the face for recognition and, equally, these descriptions can be estimated by computer vision. The new approach is akin with automated eyewitness identification, using information derived by human and automated processes. We therefore have a new approach to identifying humans by using attributes that can be described by people or by automated vision. We are then well on the way to provide a new way for automated identification from surveillance video which is a capability much needed in modern society.

16:40
Christoph Busch
Hochschule Darmstadt
Christoph Busch
Awards/ Closing Remarks

Informationen und Kontakt

Wenn Sie noch Fragen haben, wenden Sie sich bitte an:

Moderation

Christoph Busch
Tel.: +49 (0) 6151/16-30090
E-Mail:

Administration

Claudia Prediger
CAST e.V.
Tel.: +49 6151 869-230
E-Mail: