Machine Learning (core)

The aim of the course is to give to students an understanding of machine learning methods, including Neural Networks and Clustering algorithms, and to allow them to apply such methods in real world problems. Through lectures and classes that use Matlab, the students will learn several useful algorithms in machine learning and how to use them to analyse data.

Advanced Transform Methods (core)

Time-frequency transforms are an important tool in the analysis and processing of signals and images. These transforms include the Fourier transform, spectrogram, discrete cosine transform, wavelet transform, and Wigner-Ville distribution. This course aims to introduce transforms and sub-band techniques as a pre-cursor to compression and other applications. It is the first step beyond the fundamentals of Digital Signal processing.

Introduction to Computer Vision (core)

This module equips students with theoretical and empirical knowledge of computer vision principles, from foundations to applications. The module covers core concepts such as light and colour, image formation, filtering, transformations, feature extraction, motion estimation, segmentation, object representation, object detection and tracking. These concepts will be put into practice with several lab exercises. Lectures will also discuss examples of real-world computer vision applications and future prospects of this ever expanding field.

Computer Graphics (optional)

This course is concerned primarily with computer graphics systems and in particular 3D computer graphics. The course will include revision of fundamental raster algorithms such as polygon filling and quickly move onto the specification, modelling and rendering of 3D scenes. In particular, the following topics may be covered: viewing in 2D, data structures for the representation of 3D polyhedra, viewing in 3D, visibility and hidden surface algorithms, illumination computations. Some attention will be paid to human perception of colour and interactive 3D such as virtual reality.

Big Data Processing (optional)

The aim of the course is to introduce students to the paradigm of Parallel Computing, an awareness of its advantages and current limitations and allow the development of practical programming skills in a parallel computing environment. In addition, the students will familiarise with message-passing systems (MPI) as adopted by industry.

Techniques in Computer Vision (core)

Can we interact with a computer, a TV, or a game console beyond a keyboard and mouse or a remote controller, perhaps with natural gesture and/or facial expressions? Computer vision is at the heart of many such question which aims to develop methods that enable a machine to understand images and videos. This module explores more advanced topics in the area of Computer Vision, including image classification, object detection and recognition, facial analysis, motion estimation and human action recognition.

Emerging Topics in Machine Learning and Computer Vision (core)

The module aims at providing the knowledge that will enable informed choice about the models, methodologies and technologies that are needed in order to develop applications in emerging areas in the field of Computer Vision. The module will draw heavily on recent advances in the field of machine Learning, thus providing the students with solid knowledge of state-of-the-art methodologies in the domain.

C++ for Image Processing (optional)

This course gives students a practical introduction to C++ and uses this programming language to examine applications in low-level image processing. Areas covered include image representation, sampling and display, and image transforms and enhancement using point and spatial operations. Image processing methods such as convolution, frequency filtering and image restoration, compression and segmentation, are also considered.

Real Time and Critical Systems (optional)

Most computer systems do not sit on desks but are inside machines such as cars and medical devices. This module builds on undergraduate knowledge of operating systems and software engineering to introduce techniques for real-time system development in applications where the performance of the system is critical for safety. Practical work includes use of an ARM Cortex-M4 development board.

Digital Media and Social Networks (optional)

Online Social Networks (OSNs) and digital media services such as Facebook, Twitter, Flickr, YouTube are changing the way we interact with the Internet and receive our news, content and recommendations. In this module, the aim is to introduce the concepts around measurement, analysis, usability and privacy aspects of OSNs. The module will bring together a number of studies from different measurement studies on the topic, designs for new systems, and the directions that such networks are taking with the new digital media plans. This module will develop a deep understanding and analysis approach to learning specifically about Social Media and their properties.

Artificial Intelligence (optional)

The course introduces the students to techniques used in Artificial Intelligence including problem formulation, search, logic, probability and decision theory. The objective of the course is to express problems as search problems and to identify the appropriate techniques for solving them.

Real Time Digital Signal Processing (optional)

This module will cover a broad class of principles needed to develop real-time digital signal processing systems, with a particular focus on the constraints of embedded hardware. Example projects will be drawn mostly from audio, but the principles presented are equally applicable to other domains. Students will develop projects on a development kit containing an ARM Cortex-A8 processor, a platform commonly found in mobile devices.

Individual MSc Project (Dissertation)

The individual project supervised by a member of the academic staff may include industrial collaborations or commercially focused activities.

Andrea Cavallaro: Professor of Multimedia Signal Processing. He was awarded a research fellowship with BT labs in 2004; the Royal Academy of Engineering Teaching Prize in 2007; three Student Paper Awards at IEEE ICASSP in 2005, 2007, and 2009; and the Best Paper Award at IEEE AVSS 2009. He has published four books and more than 100 papers. His research interests are target tracking and multimodal content analysis for multisensor systems.

Ebroul Izquierdo: Professor of Multimedia and Computer Vision. Chair of Multimedia and Computer Vision, and head of the Multimedia and Vision Group. He holds several patents in multimedia signal processing and has published over 400 technical papers. He is a Chartered Engineer, a fellow member of the IET, a senior member of the IEEE, and a member of the BMVA.

Chrisantha Fernando: Lecturer (Assistant Professor). The title of his research is The Theory of Darwinian Neurodynamics. He now collaborates with the major experts in the field, working on a project founded by the Templeton foundation and on a 2.2M Euro European project. He is an editor for the journal PLoS ONE and a champion for bio-inspired computing on the TRUCE grant.

Shaogang Gong: Professor of Visual Computation, Fellow of the IET and of the BCS. He has published over 250 papers in computer vision and machine learning, a book on Visual Analysis of Behaviour, and a book on Dynamic Vision. His work focuses on motion and video analysis, object detection, tracking and recognition, face and expression recognition gesture and action recognition, as well as visual behaviour profiling and recognition.

Hatice Gunes: Lecturer (Assistant Professor). Her research interests are in affective computing, visual information processing, and machine learning. She worked on the European Community award winning project SEMAINE and on the project MAHNOB. She won numerous awards like Outstanding Paper at IEEE FG 2011, Quality Reviewer at IEEE ICME 2011, Best Demo at IEEE ACII 2009, and Best Student Paper at VisHCI 2006.

Miles Hansard: Lecturer (Assistant Professor). His current research is on random scene-models, geometric and statistical models of biological vision, and construction and rendering of 3D models from 'time-of-flight' range data. As part of a European project, he wrote real-time image processing and control software for the POP robot heads. He also organised the ICCV 2013 workshop on CDC4CV: Consumer Depth Cameras for Computer Vision.

Pengwei Hao: Lecturer (Assistant Professor). His research interests are in image processing and computer graphics, especially focusing on image coding and colour imaging. One of his techniques has been included in JPEG 2000, the new international standard for image coding.

Ioannis Patras: Senior Lecturer. His research is in the areas of Computer Vision and Pattern recognition with applications in Multimedia Analysis and Multimodal Human Computer Interaction. Thanks to his achievement in the field, he closely collaborates with the industry and with the top academic institutions in Europe.

Yi-Zhe Song: Lecturer (Assistant Professor). His research interests are in multimedia technology in general, but he is particularly interested in the convergence of Computer Vision, Computer Graphics and their integration into Multimedia Sensor Networks (MSN). He currently is lecture on the QM-BUPT Joint Programme, and previously lectured two undergraduate units at the University of Bath.

Tony Xiang: Senior Lecturer. His research interests include computer vision, image processing, sequential data analysis, pattern recognition, machine learning, and data mining, with publications in the most relevant conferences and journals of the respective fields. He worked on the project ICONS and INSIGHT, founded by UK government bodies.