People
Academics

Anil Kokaram received the Ph.D. degree from Cambridge University, U.K., in 1993. Then, he was a Research Fellow with the Cambridge University Engineering Department and Churchill College Cambridge. In 1998, he established the Sigmedia Group (www.sigmedia.tv) at Trinity College Dublin. In 2011 a company he founded, Green Parrot Pictures, operating in the consumer image processing space, was acquired by Google. He is currently working as a Tech. Lead in the Chrome Media Group at Google, Mountain View CA. He continues to hold his Professorship with the Department of Electronic Engineering, Trinity College Dublin and is a Fellow of the College. He is the author of over 100 refereed papers in various conferences and journals over the years. His research interests include the general area of digital video processing and algorithms for stereoscopic quality improvement and video processing in the life sciences. He is best known for his work in automated motion picture restoration.
Prof. Kokaram was an Associate Editor of IEEE Transactions in Image Processing from 2002 to 2007. He was a recipient of an Academy Award alongside colleagues at The Foundry (a visual effects software manufacturer in London) for his contribution to a range of visual algorithms in 2007. These are now widely used in the industry for motion-based effects and quality improvement (The Matrix, The Lord of the Rings, King Kong, etc.).

Dr. François Pitié is an Ussher Assistant Professor in Media Signal Processing in the School of Engineering and part of the ADAPT Research Centre. His research is concerned with media signal processing applications related to film postproduction and film restoration. The aim of his research is to develop tools that can enhance the cinematic experience of immersion. https://github.com/LCAV/pyroomacoustics/

Dr. Naomi Harte is an Associate Professor in the School of Engineering and a Fellow of Trinity College. She was appointed as an SFI Engineering Initiative Lecturer in Digital Media in 2008. Prior to returning to academia, Dr. Harte worked in high-tech start-ups in the field of DSP Systems Development, including her own company founded in 2002. She also previously worked in McMaster University in Canada. Dr. Harte’s specialist area is Human Speech Communication. Her industrial background brings a real-world approach to her research. Her work involves the design and application of mathematical algorithms to enhance or augment speech communication between humans and technology. Current projects focus on speech quality, audio-visual speech recognition, emotion in speech, speaker verification and bird species analysis. She is a Co-PI at the SFI ADAPT Research Centre.
Research fellows

Wissam A Jassim is a Research Fellow with the Sigmedia Research Group at Department of Electronic & Electrical Engineering at Trinity College Dublin, Ireland. He received his PhD degree in Electrical Engineering from University of Malaya, Malaysia, in 2012. From Oct 2019-Jan 2021, he worked as a Research Fellow with the QxLab based in the Insight Centre for Data Analytics, University College Dublin, Ireland. He joined the SFI Adapt Centre at Trinity College Dublin in 2016 and spent three years working on speech related research projects. His current research focuses on the ‘ARAMA: Using AI and Computer vision to enable low cost broadcast of Football games’ research project funded by Enterprise Ireland.

Sébastien Le Maguer is a postdoctoral research fellow based at Trinity College Dublin and funded by the IRC-GOIPD scheme. His work focuses on Speech Synthesis evaluation. Before joining ADAPT, he was a postdoctoral researcher at Saarland University/DFKI (Germany) where he worked on the integration of information density based features for speech synthesis; at INRIA/IRISA in Rennes (France) where he worked on information retrieval with application on medical context. He did his PhD on parametric speech synthesis evaluation at IRISA/Université de Rennes 1. Aside these projects, he also has interest about speech pathology, especially stutter-speech.

Justine Reverdy is a postdoctoral research fellow at Trinity College Dublin, working on RoomReader, a research project in association with UCD, within the ADAPT Centre. She recently completed her PhD in the domain of Computational Linguistics that focused on the analysis of task-based Interactions, in particular linguistic alignment phenomena. She holds a Master’s Degree in Natural Language Processing, Text and Speech, from the University Grenoble Alps. Her interests encompass multimodal, human and computer-mediated interactions, spontaneous dialogue analysis, along with the quantification of behaviours, social signals, and communication patterns.
PhD Students

Clément Bled is a PhD researcher with the Electronic Engineering department at Trinity College Dublin, Sigmedia group. His research focuses on image denoising.

Hao Wu is a PhD researcher at the SFI ADAPT research centre (personal page on the ADAPT website) and based in the Electronic Engineering department at Trinity College Dublin, Sigmedia group. Hao works on the topic of automatic generation of intra-video comments.

Daniel J. Ringis is a PhD researcher at Trinity College Dublin. He is from Trinidad and Tobago, and previously studied Electrical and Computer Engineering at the University of the West Indies, St. Augustine. His current research is focused on motion estimation and video analysis.

Samantha Kotey is a 3rd year PhD student at Trinity College Dublin funded by the Irish Research Council GOIPD scheme. She is currently exploring Multimodal learning in vision and language. Previously she attended Trinity College Dublin to receive her Masters.

Ayushi Pandey is a PhD student based at Trinity College Dublin, and is part of the D-Real CRT Programme. Her work focuses on Speech Synthesis evaluation. She has an M.A. in Cognitive Science from Johns Hopkins University, and an M.Phil in Computational Linguistics from International Institute of Information Technology, Hyderabad (IIIT-H). For her PhD project, she is exploring features in the acoustics of synthetic speech, that may contribute to the perception of various attributes in the signal, particularly its naturalness. Her previous work involves studying perception of obstruents in non-native speakers of Hindi, development of bilingual corpora in Hindi-English, and analysing the errors of forced-alignment using code-mixed speech.

Mark Anderson is a PhD student under the supervision of Dr. Naomi Harte. His research is focused on automatic monitoring of bird populations through analysis of their song. He has a background in signal processing and analogue design. He received his masters degree in electronic and electrical engineering from Trinity College Dublin in 2019.

Darren obtained his BSc in Electrical & Computer Engineering and MSc in Data Science from the University of the West Indies and is currently delving into post-processing video restoration through the exploitation of perceptual criteria!

Vibhoothi is a Research assistant under Professor Anil Kokaram. He has also started his PhD in September 2021 under Kokaram. Most of his research work is around Modern Video codecs like AV1 for optimised transcoding over the internet. He is currently working on Video Intelligence Search Platform(VISP) project which is funded by Enterprise Ireland. He is also part of various Open-source initiatives for multimedia. Xiph. Org and VideoLAN. He is also part of technical working groups of Alliance for Open-media(AOM) for standardisation of Video Encoders and Decoders. Vibhoothi received his Bachelor’s Degree in Computer Science and Engineering from Amrita Univesity(India) in 2020.
Research Assistants, interns, and other positions

Hugh Denman was not the first SigMedian to start a PhD under Prof. Kokaram, nor was he the first to finish, but of those who finished, he was the first to start (but not vice-versa). His research ranged over analysis of broadcast snooker footage, adaptive edit-point re-synchronization for retiming of dance video, shot change detection, and video segmentation. He was a member of Prof. Kokaram’s “Green Parrot Pictures” enterprise, which led to a stint at Google working within the YouTube transcoder team. He then worked in algorithmic music video generation for a few years, and is now a developer in the HFT industry. He has maintained links with SigMedia throughout, most recently assisting in the delivery of course 4C16 “Machine Learning with Media Applications”.

Sophia Rosney is a research assistant under the supervision of Anil Kokaram at Trinity College Dublin. She is currently working on Footballarama, an Enterprise Ireland-funded video processing project which aims to develop a semi-automated method of filming and broadcasting live soccer games. She graduated with a BAI and MAI in electronic and computer engineering from Trinity College Dublin, where she wrote her thesis on speech synthesis evaluation.

Louise Duquenne is an intern at SIGMEDIA. She is a Masters student at Phelma, G-INP - Institute of Engineering &Management UGA, studying signal and image processing, communication systems and multimedia. She is working in the ROOMREADER project.
Sam O’Connor Russell is a Research Assistant working under the supervision of Prof. Naomi Harte, working in the area of multimodal interaction. Sam is currently working on the RoomReader project, exploring how artificial intelligence can enhance the experience of online interaction over tools such as Microsoft Teams and Zoom. He holds an MSc in Artificial Intelligence from the University of Edinburgh, and a BAI in Electronic Engineering from Trinity College Dublin.

Ciarán Donegan is a research assistant in the Sigmedia group where he is working on an Enterprise Ireland funded project. The project aims to create a semi-automated solution to the live broadcasting of soccer matches through deep learning and image/video processing. He completed his Bachelor’s (BAI) and Master’s (MAI) degree in Electronic & Computer Engineering from Trinity College Dublin in 2020 where he specialised in video processing.
Alumni

Marco Forte is a PhD researcher with the Electronic Engineering department at Trinity College Dublin, Sigmedia group. His research focuses on foreground estimation and motion estimation for video matting. Modern deep learning techniques are leveraged to create and tune more consistent and content aware results.

Ali Karaali was born in Istanbul, Turkey and has been working as a Postdoctoral Research Fellow in Trinity College Dublin since August 2017. He holds a PhD degree from the department of Informatics, Federal University of Rio Grande do Sul, Porto Alegre Brazil.

George Sterpu received his Master’s degree in Electronics, Automation and Signal Processing from University Grenoble Alpes in 2015 and his Bachelor’s degree from University Politehnica of Bucharest in 2013. He is currently pursuing a PhD in Audio-Visual Speech Recognition under the supervision of Dr. Naomi Harte.

Joao Cabral is a research fellow at Trinity College Dublin, in the School of Computer Science and Statistics, as part of the ADAPT Centre and SIGMEDIA group. He was awarded a Ph.D. degree in Computer Science and Informatics from the University of Edinburgh, U.K., in 2010, funded by a European Commission Marie Curie Fellowship. His Ph.D. thesis contributed with the novel integration of an acoustic glottal source model in HMM-based speech synthesis, for improvement of speech quality and control over voice characteristics. Before joining Trinity College Dublin in 2013, he also worked as a postdoctoral research fellow at the University College Dublin, as part of the CNGL research centre, from 2010. His main research is on improving Text-To-Speech syntesis (TTS) methods based on statistical learning models. He is also interested in evaluation methodologies of synthetic speech in the context of multi-modal applications with virtual characters. His research interests also include analysis of emotion and affect from speech and other modalities (text, video, biosignals, etc.), voice transformation, prosody modelling, automatic speech recognition (ASR) and Natural Language Processing (NLP).