Skip to main content
Hantao Liu

Professor Hantao Liu

Professor of Human-Centric Artificial Intelligence
Director of International

School of Computer Science and Informatics

Email
LiuH35@cardiff.ac.uk
Telephone
+44 29208 76557
Campuses
Abacws, Senghennydd Road, Cathays, Cardiff, CF24 4AG
Users
Available for postgraduate supervision

Overview

NEWS:
Associate Editor
(2024-) IEEE Transactions on Image Processing - Impact factor: 10.6
Our lab developed - AGAIQA: Top-performing No-Reference Image Quality Assessment (IQA) Model (Early Access by IEEE)
Our lab developed - SSPNetFirst AI Model for Predicting Human Visual Behaviour Change (published by IEEE)
Our lab developed - Top AI Model: Predicting Radiologists' Gaze in Mammogram Reading (published by IEEE)
Our lab developed - TranSalNet: Top AI model developed for Visual Saliency Prediction (MIT benchmark|code download)
Our lab's new dataset - CUDAS: Cardiff University Distortion-Aware Saliency benchmark (published by IEEE)
Our lab's new dataset - CUID: Cardiff University Image quality Database (published by IEEE)

Grant (PI): "Omnidirectional Video Quality Assessment", funded by Royal Society (2023-)
Grant (PI): "Incoming Fellowships", funded by DFG, German Research Foundation (2023)
Grant (PI): "Human-Centric and Diversity-Aware Visual Computing for Intelligent Mobility Systems", funded by Royal Society (2022-)

Media coverage BBC: AI: Researchers train artificial intelligence to help detect breast cancer
Media coverage Fox News: AI tech aims to detect breast cancer by mimicking radiologists’ eye movements: 'A critical friend'
Invited Talk to EUVIP: Predicting Radiologists’ Gaze with Computational Saliency Models in Mammogram Reading
Invited Talk to Infection and Immunity Annual Meeting: Predicting radiologists’ gaze and decisions using deep learning

Associate Editor (2022-) for IEEE Transactions on Circuits and Systems for Video Technology, a prestigious IEEE journal dedicated to video technology

NHS honorary appointment (2022-)
Our Partnership with NHS - Breast Test Wales, University Hospital of Wales, and Great Ormond Street Hospital - has led to advancements in AI and Diagnostic Imaging. Top AI model developed for "Predicting Radiologists' Gaze in Mammogram Reading", published in a prestigious IEEE journal.

Selected current grants:

Project: Project A05 Image/Video Quality Assessment
Collaborative Research Center, SFB-TRR 161 Quantitative Methods for Visual Computing: From Test Databases to Similarity-Aware and Perceptual Dynamic Metrics. The project addresses methods for automated visual quality assessment and their validation beyond mean opinion scores. We propose to enhance the methods by including similarity awareness and predicted eye movement sequences, quantifying the perceptual viewing experience, and to apply the metrics for quality-aware media processing. Moreover, we will set up and apply media databases that are diverse in content and authentic in the distortions, in contrast to current scientific data sets.
Funded by Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)
Value: €8.34 million (Project A05 value: €490,588)
Partners: University of Konstanz (Lead)

Project: COVID-19 Vaccination for Vulnerable Namibians
There is major COVID-19 vaccination resistance in Namibia, especially amongst vulnerable and remote groups. Ministry of Health and university partners will co-produce health promotion awareness campaigns for the most disadvantaged Namibians and then deliver the vaccination programme itself, transforming tens of thousands of lives.
Funded by Welsh Government
Value: £125,000
Partners: Phoenix Project, Welsh Government

Project: Supporting the needs of Zambian patients with HIV during the COVID-19 pandemic
In Zambia HIV-positive clients receiving anti-retrovirals are regularly monitored and hence accessible beneficiaries of COVID-19 vaccination. This project will provide improved COVID-19 care for HIV-positive clients, with interventions in two provinces: indigenous language health promotion; vaccine fridge hubs and cold chain; vaccination training for healthcare workers giving antiretrovirals and Long-COVID clinics.
Funded by Welsh Government
Value: £183,000
Partners: Mothers of Africa, Phoenix Project, Welsh Government

Project: COVID-19 is Real: Making Crucial Health Information Available for All
This project will enable crucial health information to be delivered to communities and behaviour change by: Understanding perceptions of recipients; Producing tailored visual messages; Engaging communities.
Funded by Welsh Government
Value: £6,950
Partners: Mothers of Africa, Welsh Government

Biography

I am the Lead of Multimedia Computing Research Group, Cardiff University. I graduated from The University of Edinburgh, United Kingdom, and subsequently worked in the Department of Intelligent Systems at Delft University of Technology (TU Delft), The Netherlands for my PhD on Interactive Artificial Intelligence. My PhD research was funded by Philips Research Laboratories. I am a founder member of the Delft Image Quality Lab. Since 2006, I have been working closely with industry to develop next generation visual intelligence technologies. I led a project funded by Philips Research Laboratories that developed novel algorithms for visual media quality assessment; and a project funded by Philips Healthcare that addressed a number of issues related to medical imaging.

I am the Director of International for the School of Computer Science and Informatics, Cardiff University. I am a member of the School's Senior Management Team and am responsible for developing, leading and delivering the International Strategy for the School. I am the Chair of International Committee of the Centre for Artificial Intelligence, Robotics and Human-Machine Systems (IROHMS), Cardiff University.

My research interests sit at the intersection of Image and Video Processing, AI/Machine Learning, Computer Vision, Applied Perception, Multimedia Computing,  and Medical Imaging.

Academic leadership

EPSRC Associate College Member – EPSRC Peer Review College
Chair – IEEE Multimedia Communications Technical Committee, Interest Group on Quality of Experience for Multimedia Communications
Committee Member – Society for Information Display (SID), United Kingdom and Ireland Chapter
Associate Editor – IEEE Signal Processing Letters (2021-2023)
Associate Editor – IEEE Transactions on Multimedia (2017-2021)
Associate Editor – IEEE Transactions on Human-Machine Systems (2015-2021)
Associate Editor – Signal Processing: Image Communication (Elsevier) (2014-present)
Associate Editor – Neurocomputing (Elsevier) (2012-2018)
Associate Editor – Signal, Image and Video Processing (Springer) (2012-2017)
Conference Chair – IEEE International Conference on Multimedia and Expo (ICME) 2020 I British Machine Vision Conference 2019 I IEEE International Conference on Quality of Multimedia Experience 2021
Area Chair (Technical Program Committee) – IEEE International Conference on Multimedia and Expo (ICME), 2015-2017
Member (Technical Program Committee) – IEEE International Conference on Quality of Multimedia Experience (QoMEX), 2012-2019

Past Projects

Perceptually Salient Video Quality Awareness via Scene-Level Quality Assessment (funded by The Royal Society)
PI: Dr Hantao Liu
The project aims to develop technology to make any video camera automatically aware of its visual quality. In many scenarios, multiple images/videos of the same scene are captured; for example, videos of the same setting taken at different times, from different viewpoints, using different cameras, or even using the same camera with different settings. To evaluate, monitor, and optimise the system’s performance, there is a need to score/compare the images of the same scene in terms of visual quality.

Computational Models for Assessment of Diagnostic Image Quality (funded by EPSRC/GCRF)
PI: Dr Hantao Liu
The project aims to develop computational models that can automatically and reliably predict the task performance of the radiologist in the interpretation (e.g., lesion detection) of medical images. These models will be used either to support the human to augment diagnostic efficiency, or to train the human towards improved diagnostic accuracy.

Modelling Human Behavioural Responses to Distortions for Visual Quality Assessment (funded by The Royal Society)
PI: Dr Hantao Liu
Automatic visual quality assessment is the key for the optimisation of image/video acquisition, transmission, processing, and display systems. The research aims to better understand and model how the human visual system (HVS) perceives distortions in visual signals, and to develop algorithms for objective assessment of visual quality.

Medical Image Quality Assessment: Perceived Quality and Diagnostic Performance (funded by Cardiff University – KU Leuven)
PI: Dr Hantao Liu
The project aims to understand how the measured differences in image quality affect diagnostic performance, and to develop computational models that incorporate the knowledge of how radiologists understand medical images. These models will be used as valuable tools in future optimisation of medical systems and clinical procedures.

Publication

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

Articles

Book sections

Books

Conferences

Websites

Research

Research interests

My research interests sit at the intersection of Image and Video Processing, AI/Machine Learning, Computer Vision, Applied Perception, Multimedia Computing,  and Medical Imaging.

Visual Experience Computing
I am interested in how humans perceive visual information and developing computational models of visual perception. I am interested in image processing and machine vision systems that have the skills of perceptual intelligence, helping people make decisions or improving their experiences.

Perceptual Image and Video Processing
I am interested in biologically motivated visual models and integrating the perceptual elements with image and video processing algorithms. I am interested in perceptually optimised image and video engineering applications that benefit from the use of quantitative visual models.

Eye Movements and Saliency Modelling
I am interested in eye movements and computational saliency models. I am interested in integrating aspects of human visual attention with imaging and computer vision systems.

Databases

Picture Quality Databases
TUD Eye-Tracking Database
Cardiff Visual Attention and Visual Quality Toolbox

Book

Modelling Perceived Quality for Imaging Applications, 2011
Author: Hantao Liu
ISBN: 9789491211720

Teaching

Module Leader – Data Processing and Visualisation (undergraduate)
Module Leader – Human Centric Computing (postgraduate)

Biography

I am the Lead of Multimedia Computing Research Group, Cardiff University. I graduated from The University of Edinburgh, United Kingdom, and subsequently worked in the Department of Intelligent Systems at Delft University of Technology (TU Delft), The Netherlands for my PhD on Interactive Artificial Intelligence. My PhD research was funded by Philips Research Laboratories. I am a founder member of the Delft Image Quality Lab. Since 2006, I have been working closely with industry to develop next generation visual intelligence technologies. I led a project funded by Philips Research Laboratories that developed novel algorithms for visual media quality assessment; and a project funded by Philips Healthcare that addressed a number of issues related to medical imaging.

I am the Director of International for the School of Computer Science and Informatics, Cardiff University. I am a member of the School's Senior Management Team and am responsible for developing, leading and delivering the International Strategy for the School. I am the Chair of International Committee of the Centre for Artificial Intelligence, Robotics and Human-Machine Systems (IROHMS), Cardiff University.

My research interests sit at the intersection of Image and Video Processing, AI/Machine Learning, Computer Vision, Applied Perception, Multimedia Computing,  and Medical Imaging.

Specialisms

  • Artificial intelligence
  • Machine learning
  • Computer vision
  • Multimedia computing
  • medical imaging