APPENDIX 2: LETTERS IN SUPPORT OF PROPOSED ViTAB ...

somberastonishingΤεχνίτη Νοημοσύνη και Ρομποτική

13 Νοε 2013 (πριν από 3 χρόνια και 10 μήνες)

89 εμφανίσεις



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

i











Video
-
based Threat Assessment and Biometrics

ViTAB



An EPSRC Network Proposal

The CASE for SUPPORT



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

ii

Contents

1.

TRACK RECORD

1

1.1

Investigators

1

1.2

Initial Network Membership

2

2. DESCRIPTION OF TH
E PROPOSED NETWORK A
ND ITS CONTEXT

3

2.1

Purpose of the ViTAB network

3

2.2

Strategic importance for the UK

3

2.3

Impact of Network

4

2.4

Work Programme

6

2.5

Justification of Resources

8

2.6

Plans for network sustainability beyond three years

8

2.7

References

8

APP
ENDIX 1: DIAGRAMMATIC WORK PLAN

9

APPENDIX 2: LETTERS
IN SUPPORT OF PROPOS
ED VITAB NETWORK

1
0


Police Information Technology Organisation


Police Scientific Development Branch


City U
n
iversity


Queen Mary, University of London


Astraguard Ltd


Association for Biometrics


Computer Recognition Systems Ltd


Thales Research and Development
(UK) Ltd

Dr James Ferryman, Department of Computer Science, Reading University

1

Dr Graeme A. Jones, Digital Imaging Research Centre, Kingst
on University

1

Rebecca Simpson, SIRA Ltd

1

Prof M. Fairhurst, Dr F. Deravi, Department of Electronics, University of Kent

1

2.3.1

Industry and user needs

4

2.3.2

Influencing the Academic Community

5

2.3.3

The Scientific Challenges

5

2.3.4

Scope and relationship

with other academic disciplines and research networks

6

2.4.1

Aims and Objectives

6

2.4.2

Work Packages

7

2.4.3

Networ
k Operation and Management

7



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

1

1.

TRACK RECORD

1.1

Investigators

Dr James Ferryman, Department of Computer Science, Reading University

Dr Ferryman’s research interests include

model
-
based methods for people and traffic surveillance, human
-
computer interaction, robotics and autonomous systems, and "smart" cameras


areas in which he has
published 20 conference and jou
rnal papers. Dr Ferryman

s PhD

was
under

funded

the
EPSRC
Context
-
based Vision project (GR/J86322) and contributed to the Integrated Machine Vision (IMV) project
-

Model
Based Visual Surveillance (GR/K46620/01). He was an investigator on the Intelligent Cruise Control and
Collision Alert for Vehicles (SECURE) project (L
SF/E00143) with partners DERA, Lucas Industries and
Jaguar, and an investigator on a DTI/LINK project (GR/M97101/01) on vehicle classification. He is joint
investigator on two EC Framework V proposals: ADVISOR (IST
-
1999
-
11287) on people tracking in metro
s
tations, and ANFAS on modelling flood risks (IST
-
1999
-
11676). Dr James Ferryman is a reviewer for
Image and Vision Computing Journal, PAMI, the IEEE, and the Robotics and Autonomous Systems Journal.
He was the Programme Chair for the 8th International Symp
osium on Intelligent Robotic Systems, and the
IEEE International Workshops on Performance Evaluation of Tracking and Surv
eillance in 2000, 2001 and
2003, and is a rev
iewer for the

EU

Sixth Framework

IST Programme.

Dr Graeme A. Jones, Digital Imaging Research Centre, Kingston University

The Digital Imaging Research
Centre (DIRC) consists of 11 permanent academic staff and 20 researchers,
and is focused on three application areas: visual surveillance, medical imaging and post
-
production for film
and television
.
Research activity in visual surveillance has focused on t
he development of distributed
architectures and intelligent environments based on behavioural analysis for video
-
based security
applications.

Currently the Director of DIRC, Dr Graeme A. Jones

has over 15 years experience in image
and video sequence analys
is and has authored or co
-
authored over 100 technical papers related to digital
imaging and video data communications.
He was an investigator on the MAVIRIC project (ENV4
-
CT97
-
0517), which concerned visual analysis techniques applied to environme
ntal remote sensing. In 2000, Dr.
Jones chaired the British Machine Vision Association workshop on Visual Surveillance and was co
-
chair of
the IAPR Workshop on Advanced Video
-
based Surveillance Systems in 2001. He is currently co
-
investigator on an EPSRC g
rant (GR/N17706/01) on traffic simulation and optimisation on intelligent video
surveillance networks, and is investigator on the EU INMOVE project (IST
-
2000
-
37422) on intelligent
mobile video environments including visual tracking.

Rebecca Simpson, SIRA L
td

Sira Ltd is an independent not
-
for
-
profit research and technology organisation with expertise in sensing and
imaging applied to a range of industrial applications. Its activities focus both on the development of
technology, and its exploitation, working

collaboratively with users and technology suppliers in industry and
academia. Sira is the lead partner in the Imaging Faraday Partnership, a DTI/EPSRC initiative to increase
the uptake of digital imaging technology in sectors including security and survei
llance, and which currently
oversees a £12M portfolio of research activity. Rebecca Simpson is Sira’s Technical Development Manager,
with particular responsibility for setting up and managing collaborations. With a background in imaging
hardware and instru
mentation systems, she has been involved for the last five years in developing consortia
which have attracted private finance, UK national and European funding. She has expertise in project and
consortia management, and technology roadmapping.

Prof M. Fairhurst, Dr F
. Deravi, Department of Electronics, University of Kent

The University of Kent has been among the pioneers of biometrics research in the UK. The investigators in
the Network from the University of Kent will be Professor M.C. Fairhurst and Dr F. Deravi. Pro
fessor
Fairhurst was an invited speaker at the International Biometrics 2002 Conference in Amsterdam and is an
invited Keynote Speaker at the International ICDAR Conference in the summer of 2003. He was the
Academic Representative on a DTI
-
sponsored Techno
logy Mission to the USA last year to study the state of


Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

2

the art in biometrics in Law Enforcement, and is a member of the Board of the Association for Biometrics
and the UK Biometrics Working Group. Dr Deravi was a co
-
editor of the Biometrics Industries Rep
ort in
2001 published by the Elsevier Advanced Technologies. He has been an invited speaker at several
international seminars on biometrics, notably at a DTI seminar in Taiwan and as a guest speaker at CeBIT.
Both have been involved in numerous conference
s and workshops on biometrics and image classification, as
organisers, programme committee members, and presenters. In 2003 they are to be Guest Editors of a
Special Issue on Biometrics of a leading international academic journal.

1.2

Initial Net
work Membership

The network brings together the main industrial organisations and research institutions with an interest in
image
-

and video
-
based surveillance and authentication. All of the organisations listed in Tables 1 to 3 have
expressed support for

the network and willingness to participate in its activities

Organisation

Contact(s)

Contribution

SIRA Ltd

Rebecca
Simpson

Technology Provider

Police Scientific Development Branch

Richard Botten

Specification of end
-
user requirements

Police Information Technology Organisation

Ambika Suman

Specification of end
-
user requirements

Association for Biometrics

Clive Reedma
n

Specification of end
-
user requirements

CCTV Users Group

David White

Specification of end
-
user requirements

Identix (UK) Ltd

Mustafa Y. Koita

Technology Provider

Computer Recognition Systems Ltd

Bill Adaway

Technology Provider

Thales Research and Technology (UK) Ltd

Charles Attwood

Technology Provider

As
traguard Ltd

Dr Janet Jones

Technology Provider

Table
1

-

Industrial and Government Organisations


University
, Department/Centre

Contact(s)

Relevant research interests

City University

Information Engineering Centre

Dr Tim Ellis

Tr
acking, Multi
-
camera Fusion,
Activity
analysis

University of Kent

Department of Electronics

Prof Mike Fairhurst

Dr Farzin
Deravi


Biometrics and Multi
-
modal Systems,
Intelligent Biometric Interfaces

Kingston University

Digital Image Research Centre

Dr

Graeme A. Jones

Dr Sergio Velastin

Tracking, Optic Flow, Video Annotation,
Crowd Analysis,

Distributed Systems

Queen Mary, University London

Department of Computer Science

Prof Shaogang Gong

Face and Gesture, Behaviour Analysis,

Video Indexing and Search

University of Reading

Computational Vision Group

Dr James Ferryman

Tracking, Multi
-
camera CCTV, Behaviour
Analysis, Performance Evaluation

Table
2

-

Academic Membership


Organisation

Contact(s)

Contribution

Munich University of
Technology,
Institute
for

Human
-
Machine Communication

Dr Gerhard Rigoll

Object Tracking, Visual Surveillance, Face
and Gesture Recognition

University of Genoa, Italy
Department of
Biophysical and Electronic Engineering

Prof Carlo Regazzoni

Tracking, Perfo
rmance Evaluation,
Systems Integration, Scene Modelling

University of Oulu, Finland
Department of
Electrical Engineering

Professor Olli Silven

Image Sequence Analysis, 3D modelling,
Camera Calibration

Table
3

-

International Membe
rship



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

3

2.

DESCRIPTION OF THE P
ROPOSED NETWORK AND
ITS CONTEXT

2.1

Purpose of the
ViTAB

network

The
ViTAB

(
Video
-
based Threat Assessment and Biometrics
) network will specifically encourage
the
maturation of those image
-

and video
-
interpretation technologies most likely to reduce the level of crime in
town centres, public transport and sensitive sites. Emphasis will be given those technologies that will
improve the effectiveness of CCTV contr
ol room operations, or that enable identification and authentication
capability to improve on the protection of buildings and other commercial and public assets. By bringing
together academics, technology providers and police
-
service end
-
users, the network

aims to influence
research activity within the
imag
e

processing,
computer vision, machine learning and signal processing
research communities. To achieve this purpose,
ViTAB

will adopt an appropriate mix of roadmapping,
workshop, partnership facili
tation, dissemination and technology
-
transfer activities to promote a prioritised
set of capabilities guided by two of the primary police service technology experts


the Police Information
Technology Organisation (PITO) and the Police Scientific Developme
nt Branch (PSDB).

2.2

Strategic importance for the UK

There has been considerable national and international urgency in the use of identification and threat
assessment technologies to tackle crime and terrorist threats[
1
,
2
]. Currently the UK enjoys the greates
t
density of installed CCTV networks in the world particularly in
town centres and public transport systems.
However, successfully detecting crime to a large part depends on being able to establish identity, recover
evidence and, where possible, to pre
dict potential threats as they develop.

The Home Office has recently outlined a five
-
year strategic framework whose goal is “
to ensure that the
police service is equipped to exploit opportunities in science and technology to deliver effective policing

th
rough the identification of capability requirements and co
-
ordination of the technology supply chain[
2
].
This goal of identifying, evaluating and implementing appropriate enabling technology will be achieved
by
(1) establishing the
priorities

for future scientific research, (2) by
co
-
ordinating

the development of new
technology between users and suppliers, and, crucially, (3) by
anticipating

new technologies to ensure the
earliest exploitation by the police se
rvice. A key component of this strategy has been the identification of
twenty
-
seven existing or new police service

capabilities

to which new technologies are likely to make a
significant impact. Of the ten priority capabilities judged most likely to have t
he greatest impact on police
performance, four have significant requirements for
identification, authentication

and
threat assessment

through the use of biometrics or the interpretation of CCTV video streams. These are



eliminating threats to public safety
taking account of the increased risk of terrorist activity,



effective use of intelligence
-
gathering technology,



maximising the value of evidence, and



undertaking effective surveillance.

Two further priorities that could be significantly enhanced by the au
tomated annotation of stored video
streams and biometrics are



effective management of investigations including intelligent systems to assist decision making, and



authentication of identity.

Crucial to the implementation of the Home Office’s strategy on cri
me reduction will be the role of its Police
Information Technology Organisation (PITO) and the Police Scientific Development Branch (PSDB). For
both organisations, the success of the strategy will depend on their ability to focus their supply chain of end
-
users, technology providers and manufacturers around their future capability requirements. As part of this
process, the
ViTAB

network will give them the ability to influence the academic community and research
funding bodies to ens
ure the timely development of appropriate enabling technologies.



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

4

2.3

Impact of Network

2.3.1

Industry and user needs

The requirement to develop key policing and crime reduction capabilities in the areas of threat assessment,
identification and authentication are the

two primary drivers of the structu
ring activities within the ViTAB

network


see
Figure
1
. Each of these drivers can be broadly associated with the current missions of the
Police Information Technology Organisation (PITO) and the

Police Scientific Development Branch (PSDB).

Technology Providers, Manufacturers, End Users
Academia
Threat Assessment Capabilities
Identification and Authentication Capabilitie
s
PITO
PSDB
AfB
CUG
Technology
Foresight
Partnership
Facilitation
The Network
Enabling
Technologies
Capability
Foresight

Figure
1
. The
ViTAB

Network

One of the primary responsibilities of PITO is to develop national information systems in support of the
police. Of most signific
ance to this network proposal is PITO’s current commitment to its IDENT1 project
whose goal is to deliver national strategic identification solutions to the polic
e service for the future.
“PITO

s goal is to start NOW on developing the infrastructure to ens
ure that the full potential of
Identification technologies are realisable in the shortest possib
le time.”

Thus PITO fully intend

that the
ViTAB

network will enable them to locate and encou
rage the enabling technologies t
hat will un
derpin their
future capability requirements. However, to widen the scope and ensure applicability to a variety of other
end
-
users, the
Identification and Authentication

driver is also informed by the appropriate Biometrics user
-
group
i.e.

the Association f
or Biometrics (AfB). Primarily representing the technology providers, the AfB
aims to build awareness of Biometrics technologies and applications, develop and promote standards and
best practices, to identify research opportunities and promote collaborativ
e research, and to provide liaison
with national and international Government Agencies.

The Police Scientific Development Branch (PSDB) of the Home Office advises UK government and
policing organisations on the use of a wide range of CCTV and vision
-
based

detection systems for security
and policing purposes. Its brief includes the evaluation of CCTV and control room technologies including
threat assessment

systems. “Threat assessment capabilities are based on any technologies which can
improve the effectiv
eness of control room operations.” This unique expertise in control room technologies
ensures that the future
Threat Assessment

capabilities are driven by an organisation that understands the
police requirements. Nonetheless it is important to generalise t
he requirements of the criminal justice
systems to a wider pool of end
-
users. Consequently,
the
CCTV Users Group (CUG) has been enlisted as an
appropriate end
-
user organisation. Building on many years worth of operational experience, the CUG aims
to provide pr
ofessional advice for those involved in the development, management and operation of public
area CCTV systems, to encourage best practice, and to promote the effective and efficient operation and
management of CCTV systems.

Around these key end
-
users, the
ViTAB

network aims to establish an audience among technology providers,
manufacturers and the wider end
-
user community (see
Table
1
). Through
partnership facilitation
activities,
the network

will ensure
technology transfer

of nascent technologies. The
ViTAB

network will give these
end
-
users the ability to influence the academic community to ensure the timely development of appropriate
enabling technolo
gies.



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

5

2.3.2

Influencing the Academic Community

Research into video
-
based surveillance and biometrics spans a number of scientific disciplines including
signal and image processing, computer vision, audio processing, pa
ttern recognition, data mining and
distributed
processing
. Whilst academics working in these fields have naturally developed their own
informal networks, the biometrics and visual surveillance community has been somewhat fragmented.
Indeed, the international visual surveillance community has been
partitioned around two largely separate
workshop series: IEEE Workshops on Performance Evaluation of Tracking and Surveillance (
PETS2000,
PETS2001, PETS2002 and PETS
-
ICVS
2003) and the International and European workshops on Advanced
Video
-
Based Surveillance Systems (AVSS
1999, AVBS2001 and AVSS2003). The
ViTAB

network will aim
to integrate the differing foci of these communities both of which are strongly represented within the UK
(
Table
1

an
d
Table
2
) and international membership (
Table
3
) of this proposed network.

Reflecting the maturity of biometrics
-
based technologies, the Association for Biometrics and its
European
counterpart, the EU Biometric Forum, plan to publish their own roadmap[
3
] in mid
-
2003 and are co
-
ordinating with the British Standards Institute and ISO in the definition of international standards. In
contrast, the visual surveillance community i
s not as mature. Furthermore, the fragmentation
of visual
surveillance and image
-
based biometrics research has also tended to prevent the development of synergies
between the communities. It is likely that future threat assessment systems will require iden
tification of
individuals within the scene, and
it is
equally likely that authentication systems will utilise information
derived from wider CCTV based systems. The
Capability and Technology

roadmapping activities with the
ViTAB

network will provide a structured focus for these future synergies. The connectivity and reputation
of the UK visual surveillance
and biometrics
research community will ensure that this roadmap plays a
significant role in the coordination of Framework Six activit
y with the European Union.

The lack of a coherent research community also means there is a lack of any formal mechanisms for industry
(or government for that matter) to identify academic partners or identify intellectual property within the
academic commun
ity. More importantly, perhaps, is the fact that industrial providers and users are relatively
ignorant of the results of recent UK and EU funded research projects. This is particularly disappointing
given the enormous CCTV installation base and the densit
y of SME’s operating in this market segment.

2.3.3

The Scientific Challenges

There is a widespread understanding that authentication robustness and user convenience are two important
issues that could either promote or undermine the development and adoption of
biometric technologies [
4
].
The use of
multimodal

biometrics may address both these issues since a multimodal approach is likely to
improve accuracy of recognition as well as providing flexibility for the users [
5
]. The emerging complexity
requires special

management and in this regard the use of intelligent software agents is proving effective in
both user interface handling and negotiations of trust and confidence [
6
]. More generally, a number of open
issues have been identified. These include: the need f
or adaptable and learning interfaces; fusion of
environmental and non
-
biometric data; the need for interoperability and the development of standards for
multimodal plug
-
and
-
play biometrics; and evaluation


specifically effective protocols for multimodal
t
esting, incremental testing and on
-
line evaluation.

Threat assessment technologies rely on the detection and tracking of objects within the scene. Robustness to
environmental conditions and accuracy over large geographic and temporal ranges are two of the
crucial
open issues. In the USA, the ambitious VSAM (Visual Surveillance and Monitoring) project [
7
] integrated
12 research laboratories in the implementation of a people and vehicle tracking system using calibrated
cameras. Recent work in the UK has focus
sed on multi
-
camera tracking [
8
] and auto
-
calibration [
9
]. In fact
learning is likely to be a key feature of future
plug and play

surveillance components
e.g.

by observing a
large number of events to construct geometric and/or probabilistic models to suppo
rt robust tracking [
8
].
The results from object tracking systems are used for dynamic scene interpretation including behavioural
analysis. There are a number of approaches to recognising temporal scenarios based

on
probabilistic, neural
or
symbolic

networks. A state
-
of
-
the
-
art survey on approaches to learning and understanding scene activity
may be found in Buxton [
10
]. Other key issues include the analysis of scenes monitored increasingly by
mobile pan
-
tilt
-
and
-
z
oom cameras, and the monitoring of scenes with crowds
-

both present particular
difficulties to the traditional approach of event segmentation and camera calibration.



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

6

2.3.4

Scope and relationship with other academic disciplines and research networks

The proposed

network includes members with a direct involvement in many past and current UK and EU
projects involving visual surveillance and biometrics including the following EPSRC projects: Assessment
of Image Processing Techniques as a Means of Improving Personal
Security in Public Transport
GR/M29436/02; Traffic Simulation and Optimisation on an Intelligent Video Surveillance Network
GR/N17706/01; Intelligent Multi
-
Camera Surveillance and Monitoring GR/M58030; Real
-
Time Target
Identification for Security Applicati
ons GR/K44657/01; Incident Recognition for Surveillance and Security
GR/N09770/01; Automatic Vehicle Classification GR/M97107/01,Context
-
based Vision GR/J86322; Model
Based Visual Surveillance GR/K46620/01; Intelligent Agents for Multi
-
Model Biometric Iden
tification and
Control GR/N09732/01; and Enhancing Trust and Reliability in Distributed Documents GR/N37803/01. The
members have extensive direct involvement in EU projects including ADVISOR, PRISMATICA, ANFAS
and VISOR BASE, and belong to two key EU emerg
ing technology networks:
ECVision

and
FGnet
. The
membership also has extensive links with the international visual surveillance community as evidenced by
its current and previous chairmanships of the IEEE Workshops on Performance Evaluation of Tracking and

Surveillance (
PETS2000,
PETS2001, PETS2002 and PETS
-
ICVS
2003) and the International and European
workshops on Advanced Video
-
Based Surveillance Systems (AVSS1999, AVBS2001 and AVSS2003). The
network will benefit from its pre
-
existing links through SIRA Ltd to the Intel
ligent Imaging Programme
(part of the Imaging Faraday Partnership) which will devote one meeting per year to the applications of
imaging to crime prevention. The involvement of the Association for Biometrics ensures international
connectivity through the E
uropean Biometrics Forum.

2.4

Work Programme

2.4.1

Aims and Objectives


The
ViTAB

(
Video
-
based Threat Assessment and Biometrics
) network will specifically encourage the
maturation of those image
-

and video
-
interpreta
tion technologies most likely to reduce the level of crime.
This will be achieved by using the future capability requirements of police and other end
-
users to influence
the research activity of the research community. This ongoing process will aim to



Defin
e a technology roadmap based on the capability requirements of the primary end
-
users
i.e.
PITO,
PSDB, AfB and the CUG. The roadmap will identify the medium
-

and long
-
term technology needs,
the currently available intellectual property and emerging technolo
gy within the research community.



Influence the research activity of the targeted research community by setting a multidisciplinary
research agenda through the use of roadmaps, workshops, and provision of data sets, etc.



Encourage new collaboration between

academics, technology providers, manufacturers and users,
encourage the level of technology take
-
up, and enhance the research base generating this technology.

To achieve these goals of shaping the technology supply chain around the capability requiremen
ts of end
users, the following set of
specific and verifiable objectives

have been adopted:

1.

To attract a significant network membership from academics, technology providers, manufacturers and
end
-
users from all likely application domains of threat assessme
nt and identification domains.

2.

To coordinate a research agenda through the dissemination of a technology roadmap, the provision of
mechanisms for the exchange of ideas, dissemination of results, algorithms; and the provision of data
sets and ground truth a
ssociated with specific identification and threat assessment scenarios.

3.

To undertake
capability foresight

exercises whose output will be a prioritised set of image
-

and video
-
based threat assessment, identification and authentication capabilities which are

likely to a have the
most impact on crime levels.

4.

To undertake a
technology foresight

exercise for each priority policing capability to identify enabling
technologies and to identify current IP and emerging technology within the research community.



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

7

5.

To fee
dback technology awareness to the primary end
-
users to inform their own strategic planning, and
to disseminate awareness of the practical, operational and legal constraints which may be a barrier to
adoption of emerging technologies and products.

6.

To increa
se the level of connectivity between the main stakeholders within the policing and crime
reduction community, and to increase the level of synergy by disseminating awareness of current
initiatives and resources
e.g.

the current VITAL project of the PSDB.

7.

T
o enhance the level of technology take
-
up and to facilitate the creation of new collaborations between
end users, technology providers and academics through the provision of networking and presentation
sessions at workshops, and through the web
-
sites, bull
etins, etc.

2.4.2

Work Packages

The above
connectivity
,
foresight
,
agenda setting

and
technology transfer

objectives will be achieved by
implementing the following set of work packages. These are broken down into the series of work package
Tasks listed in the G
antt Chart of
Table
4
.



Capability Foresight (WP1).
The network end
-
users and end
-
user organisations are to be invited to
participate in a series of meetings to establish a prioritised list of policing capabilities to which new and
emerging technologies are likely to make the greatest impact on crime reduction. The deliverable will be
prioritised capability reports detailing the operational and legal constraints on the required technology.
This capability foresight exercise will be r
epeated
every 18 months
.



Technology Foresight (WP2)
. The foresight process will deliver a set of technology
foresight

reports
defining the technological roadmaps, scenarios and opportunities for identification, authentication and
threat assessment applicat
ions in the medium
-
term (5
-
7 years) and long
-
term (10
-
20 years). Technology
foresight
workshops

will be held every 6 months to which the whole supply chain will be invited.



Dissemination (WP3).
Establishment of a number of communications mechanisms for di
sseminating the
results of the roadmapping, to facilitate partner search and networking, and to distribute common
resources such as data sets. Mechanisms include establishment and maintenance of a website, and the
regular publication and distribution of a
bulletin. Combining the contact base of end
-
users, end
-
user
organisations, technology providers and academics, a wider directory of relevant technology providers
and academic research groups will be built to facilitate the dissemination process.



Partnershi
p Facilitation (WP4).

The identification of emerging technology, the enhancement of the
level of technology t
ake
-
up, the promotion of new consortia for funding proposals, and the creation of
new collaborative partnerships are to be objectives of this work
package. The foresight workshops are
expected to include sessions for the academic community to disseminate their work in progress and
opportunities for networking and establishment of new collaborations. Opportunities are also expected to
result from pre
-
existing links to the Intelligent Imaging Programme (Imaging Faraday Partnership).



Workshop Organisation (WP5).

Several network objectives can be addressed through the organisation
of regular workshops. These include the technology foresight exercises, pro
vision of networking
opportunities, and dissemination events for the academic community. This work package envisages a
workshop secretariat

to organise and publicise a regular series of workshops to underpin these activities.

2.4.3

Network Operation and Manageme
nt

The network will be managed by a Steering Committee formed from the Investigators and the End
-
Users and

Academic Membership. Overall responsibility for delivering the network objectives will lie with the
Principal Investigator. Responsible to the Steeri
ng Committee, Dr James Ferryman will act as the
ViTAB

Network Co
-
ordinator and coordinate the activities defined in the Work Packages.

A Project Plan defining the detailed operation and activities of the network will be tabled and

agreed at a
kick
-
off

meeting to mark the beginning of the project. As illustrated in the Gantt Chart of
Table
4
, the
Steering Committee will meet quarterly for the first year. It is expected that Membership and Wor
kshop
Registration fees will underpin the longer
-
term sustainability of the network.



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

8

2.5

Justification of Resources

Personnel:
The work of the Network Co
-
ordinator will supported by a part
-
time research assistant
responsible for the establishment and maintena
nce of the web site server and content; moderating the e
-
mail
list; assisting in the editing, production and circulation of the newsletter; preparation of materials for
workshops; and the production of the regular foresight reports. This is estimated to re
quire 25% of a full
time post over three years. Regular dedicated support will be required for the financial administration of
membership fees, sponsorship and workshop registration. Additional secretarial support for taking minutes
at workshops and foresi
ght and management meetings effort that we estimate as 10% of a full
-
time post.

Travel:

There are seven Capability and Technology Foresight activities combining meeting and workshop
events. While the costs of the workshops will be met by membership and re
gistration fees, local travel costs
of £4,200 in total is sought for the 12 Investigators, End
-
user representatives and Recognised Researchers.

Consumables:
The traditional dissemination activities anticipated in this proposal will be supplemented
with the

generation and distribution of large volumes of data sets and associated ground truth sets. This will
require the purchase of a dedicated Web and FTP server, and a DVD Duplication facility, and DVD media.

Publication Costs:

The regular Capability and For
esight activities generate a series of report deliverables
that must be professionally printed. In addition the regular network bulletin, that is to be distributed to
companies and academia, will have an important role in establish connectivity and widenin
g membership.

2.6

Plans for network sustainability beyond three years

While the Home Office strategic framework[
1
] addresses the next five years, the need to anticipate the
technological requirements of video and im
aging based analysis in future policing capabilities will be a long
term concern. Thus ongoing network activities will be desirable beyond the three
-
year duration of EPSRC
funding, and several mechanisms for supporting these activities exist. At a minimum
level, following a
successful three
-
year programme of workshops it should be possible to continue with periodic workshops
funded out of attendance fees using the established management structure. A more effective approach would
be to integrate the network
within the European research community using the EU’s Sixth Framework
Programme. Indeed, many of the members are already leading or participants in proposed EU visual
surveillance consortia
e.g.

PePe (hendrix.king.ac.uk/pepe), EVSnet (evsnet.org), European

Biometrics
Forum (eubiometricforum.com) and SURPRISE (hendrix.king.ac.uk/surprise).

2.7

References


[1]

Turning the Corner
, Foresight Crime Prevention Panel Report, Department of Trade and Industry, December. (2002)

[2]

Police Science and Technology Strategy 2
003
-
2008
, Home Office Policy Unit, January 16. (2003)

[3]

Biovision: A Roadmap for Biometrics in Europe 2003
-
2010
, European Biometrics Forum (2003)

[4]

Biometric Industry Report
-

Market and Technology Forecasts to 2003
, F Deravi and M Lockie, Elsevier Advanced
Te
chnology, ISBN 1
-
85617
-
376
-
3, December (2000)

[5]

Enhancing consensus in multiple expert decision fusion
, M C Fairhurst, IEE Proceedings of Vision, Image and Signal
Processing, 147, pp. 39
-
46 (2000)

[6]

Intelligent agents for multimodal biometrics
, International
Journal of Universal Access in the Information Society, M C
Fairhurst, R Guest, A Canuto and N Mavity. To appear (2003)

[7]

A System for Video Surveillance and Monitoring
, R. Collins, Lipton, Kanade, Fujiyoshi, Duggins, Tsin, Toliver, Enomoto and
Hasegawa, VSA
M Final Report, CMU
-
RI
-
TR
-
00
-
12, Robotics Institute, Carnegie Mellon University, May (2000)

[8]

Multi
-
Camera Image Tracking
, J. Black and T. Ellis, In Proceedings 2nd IEEE International Workshop on Performance
Evaluation of Tracking and Surveillance, Kauai, Ha
waii, December 9 (2001)

[9]

Auto
-
Calibration in Multiple
-
Camera Surveillance Environments
, G.A. Jones, J. Renno, and P. Remagnino, Third IEEE
International Workshop on Performance Evaluation of Tracking and Surveillance, Copenhagen, June 2002. IEEE Press (2002
)

[10]

Generative Models fort Learning and Understanding Scene Activity
, H. Buxton, Technical Report DIKU
-
TR
-
2002/01 (ISSN
0107
-
8283), The University of Copenhagen, pp 71
-
81, June (2002)



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

9

APPENDIX 1: DIAGRAMMATIC WORK PLAN


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Task 1.1
First Capability Foresight Exercise
M4
Task 1.2
Second Capability Foresight Exercise
M8
Task 2.1
Technology Foresight Exercise
M6
Task 2.2
Technology Foresight Exercise
M7
Task 2.3
Technology Foresight Exercise
M9
Task 2.4
Technology Foresight Exercise
M10
Task 2.5
Technology Foresight Exercise
M12
Task 3.1
Establish Website
M2
Task 3.2
Establish Bulletin
M3
Task 3.3
Compile Distribution Lists
M5
Task 3.4
Disseminate Foresight Reports
Task 4.1
Partnership Exercise
Task 4.2
Partnership Exercise
Task 4.3
Partnership Exercise
Task 4.4
Partnership Exercise
Task 4.5
Partnership Exercise
Task 5.1
Organise Network Workshop
Task 5.2
Organise Network Workshop
Task 5.3
Organise Network Workshop
Task 5.4
Organise Network Workshop
Task 5.5
Organise Network Workshop
Task 6.1
Kick-Off Meeting and Project Plan
M1
Task 6.2
Quarterly Steering Committee Meetings
Task 6.3
Develop Continuation Plan
M11
Task 6.4
Coordinate Final Report
M13
WORKSHOPS
WK1
WK2
WK3
WK4
WK5
WP5 Workshop Organisation
WP6 Management and Coordination
WP4 Partnership Facilitation
Work Packages
WP1 Capability Foresight
WP2 Technology Foresight
WP3 Dissemination

Table
4

Gantt Chart

Video
-
based Threat Assessment and
Biometrics

(
ViTAB
)

Network

10







Work Plan

APPENDIX 2: LETTERS
IN SUPPORT OF PROPOS
ED
VITAB

NETWORK



Video
-
based Threat Assessment and Biometrics

(
ViTAB
)

Network

9

Variables
A

Video
-
based Threat Assessment and Biometrics

Variables
B

ViTAB