Project suggestions from the Graphics & Interaction Group
Real-Time Gaze Estimation and Eye Tracking on an Android Portable Device
Originator: Andreas Bulling
Researchers in human-computer interaction as well as hardware manufacturers actively investigate new ways of interacting with portable devices. Latest devices equipped with high-resolution front-facing cameras promise unobtrusive gaze-based interfaces that will become pervasively usable in everyday life. The goal of the current project is to develop and evaluate algorithms for robust gaze estimation and real-time eye movement tracking on an Android portable device. The project builds on a previous project on eye gesture recognition on mobile phones. The project will involve (1) the collection of a large corpus of eye images recorded using a portable device's front-facing camera, (2) the development of low-level image features to estimate discrete gaze directions, and (3) the development and evaluation of algorithms for continuous and real-time eye tracking in a user study.
Requirements: Basic knowledge of computer vision, probabilistic modelling and Android development beneficial
The project might be appropriate for a Part II project, if the student has (some of) the above skills, but is more suitable as a MPhil project.
Multimodal Interaction on an Android Portable Device
Originator: Andreas Bulling
Touch has emerged as the predominant modality for interacting with portable devices, such as mobile phones or tablets. The advent of powerful devices equipped with high-resolution front-facing cameras paves the way for gaze to complement touch and to provide more powerful and natural styles of interaction. The goal of the current project is to develop and evaluate an interface for simultaneous touch- and gaze-based input on an Android portable device. The project builds on a previous project on eye gesture recognition on mobile phones. The project will involve (1) the identification of common tasks performed on portable devices and their adaptation for multimodal interaction, (2) the development of a multimodal interface that implements these tasks, and (3) the development of a realistic experimental scenario and the comparison of multimodal interaction with unimodal eye gesture- and touch-based approaches in a user study.
Requirements: Basic knowledge of Android development, computer vision and interaction design beneficial
The project might be appropriate for a Part II project, if the student has (some of) the above skills, but is more suitable as a MPhil project.
One Day in the Life of the Eyes
Originator: Andreas Bulling
Research on eye-based context recognition has demonstrated that a variety of visual and non-visual activities can be recognised from eye movements. Most work to date has focused on stationary settings and short-term activities typically performed on a desktop computer, such as reading, typing, browsing or watching a video. The goal of the current project is to develop and evaluate algorithms for associating eye movements recorded in mobile daily life settings with longer-term activities or situations. Following an initial data acquisition, the first step will be to analyse the eye movement data and to develop features that are suitable for detecting such activity changes or situations. These features may for example cover basic eye movement characteristics or higher-level behavioural patterns. Afterwards, algorithms for feature extraction and eye movement analysis will be implemented to achieve an entire automatic detection process. The system will be evaluated in a long-term user study.
Requirements: Basic knowledge of signal processing, machine learning and MATLAB/C++ beneficial
The project might be appropriate for a Part II project, if the student has (some of) the above skills, but is more suitable as a MPhil project.
Automatic Assessment of Road Surface Quality Using Participatory Sensing
Originator: Andreas Bulling
In 2006, the road network in Great Britain consisted of nearly 250.000 miles of roads and motorways. Monitoring such a large amount of road surface for potholes and similar types of damage is an extremely cumbersome and time-consuming task. In particular those cycling to work will know from their own experience that prompt repair often fails. One of the reasons for this is the limited number of people continuously monitoring the road surface quality. This project investigates participatory sensing for assessing road surface quality. More specifically, the goal is to develop an Android application that uses sensors readily available in current smartphones - such as the accelerometer, gyroscope, magnetometer, and GPS - and algorithms for automatic signal processing and pattern recognition to identify those (parts of) roads that need attention most urgently. As an incentive for users the application should also calculate additional information, such as number of miles cycled, top/average speed etc. and the possibility to publish all of this information online.
Requirements: Previous experience with Java/Android programming, signal processing and pattern recognition beneficial
The project might be appropriate for a Part II project, if the student has (some of) the above skills, but is more suitable as a MPhil project.
Control room attention models
Originators: Peter Robinson
Many complex industrial systems are run from central control rooms where operators monitor information on multiple screens to identify anomalous conditions. Current design tools for control rooms are limited to 3D models of the hardware which can be used to assess the physical ergonomics, but do not help understand the work of human operators.
This project focuses on developing computational models for predicting the operators' attention so that the human-machine interface could be evaluated and configured properly during control room design. These models are expected to improve arrangement of information shown through the HMI and lessen the operators' risk of missing important information in critical situations. This will involve predicting visual search patterns over an extended display.
Variants on Doo-Sabin subdivision
Originator: Neil Dodgson
Our recent paper, Deriving box-spline subdivision surfaces, describes five dual quadrilateral binary box-spline schemes (Fig. 8) that can all be considered variations of Doo-Sabin subdivision. Two of those five have never been extended to the extraordinary cases (the C2 six-arrow scheme and the C3 eight-arrow scheme). The project is to extend both of these schemes to handle extraordinary cases, then to implement all five schemes and compare them against one another.
Our paper also suggests a couple of other new subdivision schemes, that may be worth investigating as an alternative project. These are described in Sect. 7, bullet points 2 and 3.
This project requires good mathematical knowledge, facility with handling large data structures, and an understanding of 3D graphical presentation and interaction. It might be appropriate for a Part II project, if the student has excellent mathematical skills, but is more suitable as an MPhil project.
Beyond Catmull-Clark subdivision
Originator: Jiří Kosinka
Catmull-Clark is a recursive subdivision scheme that produces smooth curved surfaces from a polyhedral mesh. At each subdivision step, simple rules are used to create a finer mesh. The process terminates when polygons are considered small enough (about pixel-sized in graphics applications).
In this project, we re-implement Catmull-Clark subdivision using stencils to create the vertices of the new mesh. This would immediately allow generalising it to non-uniform knots, something we would like in our current EPSRC-funded research. Then, incorporate `averaging' to produce tensor-product stencils so that unequal knot-intervals are supported. Some tuning and tweaking at EVs would be required. Compare results with original Catmull-Clark subdivision and other subdivision surfaces. [One could proceed similarly with Doo-Sabin subdivision.]
This project requires good mathematical knowledge, facility with handling large data structures, and an understanding of 3D graphical presentation and interaction. It is suitable as an MPhil project but is too large for a Part II project.
Splines with semi-sharp features
Originator: Jiří Kosinka
Splines are usually designed using control points. The idea behind this project is to generalise this concept to support control vectors as well. This approach, if implemented using subdivision, would allow intuitive modelling with semi-sharp features.
This project is scalable (implementation can support either curves only, or both curves and surfaces) and thus it is suitable as a Part II or an MPhil project.
Hayes surfaces via subdivision
Originator: Jiří Kosinka
The original idea of Hayes was to generalise B-spline surfaces by allowing knots to be functions of the parameter. The project is to reimplement and extend this idea using subdivision. Compared to standard B-spline surfaces, this would give rise to a more flexible modelling tool within a multi-resolution framework.
This project might be appropriate for a Part II project, if the student has excellent mathematical skills, but is more suitable as an MPhil project.
Dramatic decimation for curves
Originator: Jiří Kosinka
Our current EPSRC-funded research focuses on dramatic decimation for surfaces: given a dense triangular mesh, extract a good sparse control mesh that captures the main features of the input. Applications range from multi-resolution modelling to data compression.
However, for open meshes, the boundary needs to be treated separately. The project is: given a curve (dense polygon), find a good sparse representation for it. One can look at singularities, inflections, other special points, or use Morse theory (with curvature). B-spline subdivision (possible with features) can then be used to reconstruct the input curve.
This project is suitable for a Part II project or an MPhil project.
Driving through Google street view
Originators: Peter Robinson
Google street view presents an environment as a series of still pictures, with an awkward user interface. It would be better if the view changed continuously and could be controlled from a driving simulator.
This first part of the project would be build a rendering system that takes a sequence of images from Google street view and interpolates smoothly between them. This would also involve transforming the images to a consistent frame and handling the images of other vehicles in some way. The second part of the project would be to provide better control, and would involve pre-fetching of images for a variety of paths through the data to give smooth real-time performance.
Facial expression synthesis
Originator: Tadas Baltrušaitis
Facial expressions and head gestures are some of the most important affective (emotional) cues in human interaction. The purpose of the project is to analyse the importance of head pose and facial expression separately and their interaction when expressing emotion/affect. This will aid the design of automatic affect recognition systems and allow for a better understanding of human behaviour. The project will involve (1) tracking facial expressions and head pose (most of the software needed for this is already available through Active Appearance Model and Constrained Local Model libraries), (2) synthesising the tracked expression and pose separately to create new videos (datasets of naturalistic emotions/mental states are available, although the student might opt to record new ones), (3) running a user study on the videos to see the differences in emotion recognition results on different videos. The experiment would be somewhat similar in spirit (although simpler) to the one performed by Boker et al.
Requirements: This project would benefit from previous experience with computer vision. Understanding of statistics would also be helpful. It is suitable as an MPhil project but is too large for a Part II project.
Propose your own project
The Graphics & Interaction Group has a range of interesting hardware. Consider the useful research that could be done if you had access to this and propose something novel and interesting.