Browsing by Author "Colreavy-Donnelly, S."

Now showing 1 - 8 of 8
  • Thumbnail Image
    ItemOpen Access
    Assessing the Perceived Realism of Agent Grouping Dynamics for Adaptation and Simulation
    (Elsevier, 2019-10-14) O'Connor, S.; Shuttleworth, J.; Colreavy-Donnelly, S.; Liarokapis, F.
    Virtual crowds are a prominent feature for a range of applications; from simulations for cultural heritage, to interactive elements in video games. A body of existing research seeks to develop and improve algorithms for crowd simulation, typically with a goal of achieving more realistic behaviours. For applications targeting human interaction however, what is judged as realistic crowd behaviour can be subjective, leading to situations where actual crowd data is not always perceived to be more real than simulation, making it difficult to identify a ground truth. We present a novel method using psychophysics to assess the perceived realism of behavioural features with respect to virtual crowds. In this instance, a focus is given to the grouping dynamics feature, whereby crowd composition in terms of group frequency and density is evaluated through thirty-six conditions based on crowd data captured from three pedestrianised real-world locations. The study, conducted with seventy-eight healthy participants, allowed for the calculation of perceptual thresholds, with configurations identified that appear most real to human viewers. The majority of these configurations correlate with the values extracted from the crowd data, with results suggesting that viewers have more perceptual flexibility when group frequency and density are increased, rather than decreased.
  • Thumbnail Image
    ItemOpen Access
    Efficient Computation of the Nonlinear Schrödinger Equation with Time-Dependent Coefficients
    (MDPI, 2020-03-07) Kosti, Athinoula A. ; Colreavy-Donnelly, S.; Caraffini, Fabio; Anastassi, Zacharias
    Motivated by the limited work performed on the development of computational techniques for solving the nonlinear Schrödinger equation with time-dependent coefficients, we develop a modified Runge-Kutta pair with improved periodicity and stability characteristics. Additionally, we develop a modified step size control algorithm, which increases the efficiency of our pair and all other pairs included in the numerical experiments. The numerical results on the nonlinear Schrödinger equation with periodic solution verified the superiority of the new algorithm in terms of efficiency. The new method also presents a good behaviour of the maximum absolute error and the global norm in time, even after a high number of oscillations.
  • Thumbnail Image
    ItemOpen Access
    EmotiMask: Mapping Mouth Movements to an LED Matrix for Improving Recognition When Teaching With a Face Mask
    (International Academic Forum, 2022-09-27) Hasshu, S.; O'Connor, S.; Colreavy-Donnelly, S.; Kuhn, S.; Caraffini, F.
    The Covid-19 pandemic has led to the adoption of face masks in physical teaching spaces across the world. This has in-turn presented a number of challenges for practitioners in the face-to-face delivery of content and in effectively engaging learners in practical settings, where face coverings are an ongoing requirement. Being unable to identify the mouth movements of a speaker due to the lower portion of the face being obscured can lead to issues in clarity, attention, emotional recognition, and trust attribution, negatively affecting the learning experience. This is further exacerbated for those who require specialist support and those with impairments, particularly those centred around hearing. EmotiMask embeds an LED matrix within a face mask to replicate mouth movements and emotional state through speech detection and intelligent processing. By cycling through different LED configurations, the matrix can approximate speech in-progress, as well as various mouth patterns linked to distinct emotional states. An initial study placed EmotiMask within a HE practical session containing 10 students, with results suggesting a positive effect on clarity and emotional recognition over typical face masks. Further feedback noted that it was easier to identify the current speaker with EmotiMask, however speech amplification, additional led configurations, and improved portability are desired points of refinement. This study represents a step towards a ubiquitous solution for tackling some of the challenges presented when teaching in a pandemic or similar situations where face coverings are a requirement and has potential value in other sectors where such scenarios present themselves.
  • Thumbnail Image
    ItemOpen Access
    I-Ulysses: A Technical Report
    (Elsevier, 2019-10-31) Colreavy-Donnelly, S.; O'Connor, S.; Homapour, E.
    The I-Ulysses: Poetry in Motion project is as a virtual reality experience, guiding the user through the unfolding events of James Joyce's Ulysses in real-time. Currently there is a lack of research looking at adapting literature into virtual reality, for the purposes of cultural heritage, or for serious learning. I-Ulysses is aimed at addressing this gap, by providing an educational tool, intended to help the user understand key aspects of the book. What follows is a technical report, outlining the objectives, aims and results of the I-Ulysses project.
  • Thumbnail Image
    ItemOpen Access
    Identifying Parkinson’s Disease Through the Classification of Audio Recording Data
    (IEEE, 2020-07) Bielby, James; Kuhn, Stefan; Colreavy-Donnelly, S.; Caraffini, Fabio; O'Connor, S.; Anastassi, Zacharias
    Developments in artificial intelligence can be leveraged to support the diagnosis of degenerative disorders, such as epilepsy and Parkinson’s disease. This study aims to provide a software solution, focused initially towards Parkinson’s disease, which can positively impact medical practice surrounding degenerative diagnoses. Through the use of a dataset containing numerical data representing acoustic features extracted from an audio recording of an individual, it is determined if a neural approach can provide an improvement over previous results in the area. This is achieved through the implementation of a feedforward neural network and a layer recurrent neural network. By comparison with the state-of-the-art, a Bayesian approach providing a classification accuracy benchmark of 87.1%, it is found that the implemented neural networks are capable of average accuracy of 96%, highlighting improved accuracy for the classification process. The solution is capable of supporting the diagnosis of Parkinson’s disease in an advisory capacity and is envisioned to inform the process of referral through general practice.
  • Thumbnail Image
    ItemOpen Access
    An integrated approach for mixture analysis using MS and NMR techniques
    (Royal Society of Chemistry, 2019-02-08) Kuhn, Stefan; Colreavy-Donnelly, S.; Santana De Souza, J.; Moreira Borges, R.
    We suggest an improved software pipeline for mixture analysis. The improvements include combining tandem MS and 2D NMR data for a reliable identification of its constituents in an algorithm based on network analysis aiming for a robust and reliable identification routine. An important part of this pipeline is the use of open-data repositories, although it is not totally reliant on them. The NMR identification step emphasizes robustness and is less sensitive towards changes in data acquisition and processing than existing methods. The process starts with a LC-ESI-MSMS based molecular network dereplication using data from the GNPS collaborative collection. We identify closely related structures by propagating structure elucidation through edges in the network. Those identified compounds are added on top of a candidate list for the following NMR filtering method that predicts HSQC and HMBC NMR data. The similarity of the predicted spectra of the set of closely related structures to the measured spectra of the mixture sample is taken as one indication of the most likely candidates for its compounds. The other indication is the match of the spectra to clusters built by a network analysis from the spectra of the mixture. The sensitivity gap between NMR and MS is anticipated and it will be reflected naturally by the eventual identification of fewer compounds, but with a higher confidence level, after the NMR analysis step. The contributions of the paper are an algorithm combining MS and NMR spectroscopy and a robust nJCH network analysis to explore the complementary aspect of both techniques. This delivers good results even if a perfect computational separation of the compounds in the mixture is not possible. All the scripts will be made available online for users to aid studies such as with plants, marine organisms, and microorganism natural product chemistry and metabolomics as those are the driving force for this project.
  • Thumbnail Image
    ItemOpen Access
    A Neural Network for Interpolating Light-Sources
    (IEEE, 2020-07-13) Colreavy-Donnelly, S.; Kuhn, Stefan; Caraffini, Fabio; O'Connor, S.; Anastassi, Zacharias; Coupland, Simon
    This study combines two novel deterministic methods with a Convolutional Neural Network to develop a machine learning method that is aware of directionality of light in images. The first method detects shadows in terrestrial images by using a sliding-window algorithm that extracts specific hue and value features in an image. The second method interpolates light-sources by utilising a line-algorithm, which detects the direction of light sources in the image. Both of these methods are single-image solutions and employ deterministic methods to calculate the values from the image alone, without the need for illumination-models. They extract real-time geometry from the light source in an image, rather than mapping an illumination-model onto the image, which are the only models used today. Finally, those outputs are used to train a Convolutional Neural Network. This displays greater accuracy than previous methods for shadow detection and can predict light source-direction and thus orientation accurately, which is a considerable innovation for an unsupervised CNN. It is significantly faster than the deterministic methods. We also present a reference dataset for the problem of shadow and light direction detection. © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
  • Thumbnail Image
    ItemOpen Access
    Shallow Buried Improvised Explosive Device Detection Via Convolutional Neural Networks
    (IOS Press, 2020-07-03) Colreavy-Donnelly, S.; Caraffini, Fabio; Kuhn, Stefan; Gongora, Mario Augusto; Florez-Lozano, Johana; Parra, Carlos
    The issue of detecting improvised explosive devices, henceforth IEDs, in rural or built-up urban environments is a persistent and serious concern for governments in the developing world. In many cases, such devices are plastic, or varied metallic objects containing rudimentary explosives, which are not visible to the naked eye and are difficult to detect autonomously. The most effective strategy for detecting land mines also happens to be the most dangerous. This paper intends to leverage the use of a Convolutional Neural Network (CNN) to aid in the discovery of such IEDs. As part of a related project, an autonomous sensor array was used to detect the devices in terrains too hazardous for a human to survey. This paper presents a CNN and its training methodology, suitable to make use of the sensor system. This convolutional neural network can accurately distinguish between a potential IED and surrounding undergrowth and natural features of the environment in real-time. The training methodology enabled the CNN to successfully recognise the IEDs with an accuracy of 98.7%, in well-lit conditions. The results are evaluated against other convolutional neural systems as well as against a deterministic algorithm, showing that the proposed CNN outperforms its competitors including the deterministic method.