Anonymity in P2P
networks
This project will focus on the various techniques employed in anonymous
P2P computer networks. The various attacks on these networks and the
technical drawbacks they face. Moreover we will investigate different
approaches in strengthening the existing anonymous networks. Networks
to be investigated are: MUTE, ANTS, Waste and many more.
|
Barcode and 2D codes
for mobile phones
The aim of the project is to develop an application that employs the
build-in camera of a mobile phone for reading barcodes or 2D codes from
products and identifying the individual codes. Possible application
scenarios include m-commerce, e-ticketing, mobile couponing etc.
|
Context-aware
information retrieval
Context-aware information retrieval aims to improve the usefulness of
information delivered to a user by taking into account the context in
which the search is taking place. Context here refers both to the user
operating the system and environmental meta data, such as location or
time. We are particularly interested in retrieval in mobile computing
environments where users require reliable data and are unlikely to wish
to engage in extensive document browsing.
|
Content-based
image database
Collections of digital imagery are widely used in various aspects of
life. Examples include, stock photo libraries, movie libraries,
satellite imageries etc. Research in information retrieval, database
management, computer vision and image processing is converging towards
providing content-based access to such image collections.
The fundamental issues in the design of image databases revolve around
effective and efficient selection of images which satisfy a user's
query. The major issues facing content-based retrieval are: data
modeling, feature extraction, selecting appropriate features for
content representation, query languages and indexing techniques. To
provide effective retrieval, we need to find which features are
significant in a particular domain and how these features can be used
to search for images of interest. Moreover, effective ways of
representing image data, efficient query languages and indexing
techniques are needed to enable fast search of the databases.
|
Digital
video watermarking
The aim of the project is to develop a robust video watermarking system
that could be used for broadcast monitoring. Watermarking enables us to
hide invisible data in the digital video. This data can be used for
tracking, fingerprinting, copyright infringement detection or any other
application that requires some hidden data.
|
Distributed
web retrieval
In distributed information retrieval a single query is sent to a number
of separate search engines. The documents returned by these separate
engines must then be merged into a single list for presentation to the
user. The key research challenge is how to combine these lists
optimally to deliver maximum retrieval accuracy to the user.
|
Driver
assistance using video processing
In this project we assume that there is a camera mounted on a vehicle
and it points towards the road. We want a real time video analysis
system that will track traffic signs (for example speed limit signs,
warning signs), other vehicles that are either in front of our car and
in the same lane as we are or in the opposite lane, the lanes on the
road so that we can determine if the car is slipping away from the road
and finally objects like animals, humans etc that are beside the road
and are moving towards the road and that might interrupt our vehicles
root.
|
Encryption
over mobile phones
In this projects we will investigate the possibility of developing a
Java application that will encrypt the communication between two mobile
phones using public key/private key cryptography. The application will
be running on both phones. This way the communication between Bob and
Alice will be confidential and out of the reach for Big Brother.
|
EM-based
image region segmentation for retrieval
Automatically dividing an image into a number of arbitrarily-shaped
sub-regions. This is normally referred to as the process of
region-based image segmentation and many different techniques exist.
|
Human
body and face detection in movies
The detection of bodies and faces in images or video clips is an
essential tool for many applications in fields such as security or
advanced computer/human interaction. Furthermore, it is a useful
pre-processing step for analysis tasks such as event detection and
major cast identification in broadcast content. Although it is a
complex problem there are a number of existing solutions available.
This project will involve evaluating some of these solutions in the
context of a larger scale event detection system currently under
development, in order to assist in the task of dialogue detection in
movies.
|
Gaze
tracking
Gaze information plays an important role in identifying a person's
focus of attention. The information can provide useful communication
cues to a multi modal interface. For example, it can be used to
identify where a person is looking, and what he/she is paying attention
to. In this thesis we will investigate the usage of a simple PC camera
(web camera) in order to identify where the user is looking on the
computer screen. With eye-tracking, a computer can precisely estimate
where a person is looking.
|
Image
search engine
The goal of this project is to build an image search engine. However
this image engine will not be simply a reworking of Google's image
search system. Rather it will take the text search facility of Google,
extend that by including low-level feature matching for the images
(e.g. color).
|
MPEG-4
object-based encoding
MPEG-4 is the last in a series of Multimedia content encoders that were
standardized during the 1990s. It is important because it supports the
coding of objects within a video frame. The versions of MPEG-4 encoders
and players commercially available do not support this Video Object
mode, so a new version is needed.
|
Music
discovery engine
This project calls for the creation on a multimedia database
application and engine for finding audio data files (or segments of
digital audio) that sound similar to given sounds, or that sound
similar to predefined classes of sounds. For example, multimedia
databases can easily have thousands of audio recordings. Such libraries
often are poorly indexed and to search for a particular sound or class
of sound (e.g., applause or music or the speech of a particular
speaker) can be a daunting task. To deal with the previous problem we
need to develop an application that uses Accoustic Fingerprints to
semi-automatically identify tracks in music collections.
|
Obstacle
avoidance system
The main objective of this project is to develop a system that uses two
linear CCD censors (cameras) to detect obstacles that are blocking the
path of a autonomous airplane. Since the airplane under consideration
(UAV) is lightweight the entire implementation calls for a very light
system. The cameras should be light, connect through RS232 and have
small physical dimensions.
|
Region-based
image watermarking
The main objective here is to embed and detect a watermark in the
digital spatial image domain, based on segmenting the image and
locating regions that are robust to several image manipulations.
|
SMS-TB:
An SMS with limited life span
Main objective of this project is to develop the required technology
for SMS that have a limited time to live (TTL). This SMS might contain
user-sensitive or time-sensitive material. By controlling the time that
this SMS is allowed to live in the receiver's mobile before securely
deleting it we improve the over security in e-banking and e-commerce
scenarios.
|
SMS
and user behavior modeling
When a mobile phone is stolen and an SMS is send then the receiver of
the message might be tricked into reveal sensitive information to the
imposture. Our objective is to model the user's habits and through this
modeling to determine abnormal patterns in the behavior. In case an
abnormality is determined then and extra security check will safe-guard
any sensitive information.
|
The
wayback machine for TV news
This project will aim to put each news story in its context by locating
previous stories from this news item and linking them together in a
time line. By using the teletext it is possible to construct a time
graphic progression through the news story. The teletext subtitles in
the news broadcast are not well aligned with the audio signal since it
is often broadcast a variable number of seconds after the corresponding
audio. This means that the text cannot be used for intelligent
processing of the audio-video signal such as automatically segmenting
it into individual stories. The objective of this project is to align
the text with the words spoken into the audio soundtrack.
|
Text
Mining - Tell me about ...
The aim of this project is to involve in the query process of features
such as automatically linking documents with related content and
indicating these links to the user. For example, if the user wants to
find out more about a specific topic within a document, they should be
directed straight to potentially useful additional documents.
|
Video
summarization
Browsing multimedia data to locate particular information can be an
inefficient and time-consuming process. In the simplest solution the
user can view the material from the beginning. The aim of this project
is to develop a video summarization system for transcription and key
frames. This will involve use of text summarization techniques, but
also potentially information image analysis of the key frames to decide
which parts of the text and key frames to be shown to the user. This
could be extended to showing excerpts from the complete video or to
develop the tool to expand the summary in relevant sections or link
sections of related stories.
|
Video
shot cut detection
A video shot is a single piece of continuous video shot with a single
camera. A video shot cut is an abrupt or gradual change from one video
shot to another. Detecting video shot cuts is a key starting point for
the semantic analysis of video.
|
Video
copy identification
This project requires the creation of video "fingerprints" for a set of
TV commercials. Fingerprints will be based on automatically detected
audio and video features such as color, texture, motion, pitch, etc.
For each commercial in a "database" we will generate a fingerprint and
then match these fingerprints against a live TV station transmission.
This will allow us to determine which commercials are broadcast, when
and how often.
|