Paul Rosin

Professor of Computer Vision
School of Computer Science & Informatics
Cardiff University

[ Some History | Research Interests | Software | Demos | Data | Publications | Contact Details ]



Some History

I gained the B.Sc. degree in Computer Science and Microprocessor Systems in 1984 from Strathclyde University, Glasgow, and the Ph.D. degree in Information Engineering from City University, London in 1988. I was a research fellow at City University, developing a prototype system for the Home Office to detect and classify intruders in image sequences. I worked on the Alvey project "Model-Based Interpretation of Radiological Images" at Guy's Hospital, London before becoming a lecturer at the Department of Computer Science, Curtin University of Technology, Perth, Australia. I moved to Italy to work at the Institute for Remote Sensing Applications at the Joint Research Centre. followed by a return to the UK, becoming lecturer at the Department of Information Systems and Computing, Brunel University, London, UK. At the beginning of 2000 I moved to the School of Computer Science & Informatics, Cardiff University.


Research Interests

I am a member of the School's Visual Computing research group. Some of its members are: . As well as working with my Visual Computing group colleagues I have also collaborated with other researchers such as: . We have a weekly "vlunch" seminar series which I have organised for many years. For historic interest I have listed some of the hundreds of previous talks since 2003.

My past and present research includes:
cellular automata diatoms art and architecture CBIR mesh filtering reliefs remote sensing NPR scrolls bas relief

For a number of years I was newsletter editor and then secretary of the British Machine Vision Association.

Some of my other research related activities over the years are shown below.

remote sensing logo I was co-guest editor of the special issue of Remote Sensing on "Advances and Challenges on Multisource Remote Sensing Image Fusion: Datasets, New Technologies, and Applications".
CAD Graphics I co-organised the 2023 International Conference on Computer-Aided Design and Computer Graphics.

Dagstuhl Seminar

I co-organised the 2023 Dagstuhl Seminar Media Forensics and the Challenge of Big Data.
CVM2022

Computational Visual Media

I co-organised the 2022 Computational Visual Media Conference (CVM).
SHREC21

Shape Complexity

I co-organised the SHREC'21 competition Quantifying Shape Complexity which was held at the 3D Object Retrieval workshop (3DOR'21).
3DOR21

3D Object Retrieval

I co-organised the 3D Object Retrieval workshop (3DOR'21).

NPRportrait (version 1)

The 60 images that make up the NPRportrait benchmark (version 1.0) are available.
cover

3D Imaging, Analysis and Applications (2nd edition)

Along with Yonghuai Liu, Nick Pears and Patrik Huber I edited a book on 3D Imaging, Analysis and Applications (2nd edition) which was published by Springer in 2020.

Shape Correspondence with Non-Isometric Deformations

I co-organised the SHREC'20 competition Shape Correspondence with Non-Isometric Deformations which was held at the Workshop on 3D Object Retrieval (3DOR)

Non-Rigid Shape Correspondence of Physically Based Deformations

I co-organised the SHREC'20 competition Non-Rigid Shape Correspondence of Physically Based Deformations which was held at the Workshop on 3D Object Retrieval (3DOR)
cover

RGB-D Image Analysis and Processing

Along with Yu-Kun Lai, Ling Shao and Yonghuai Liu I edited a book on RGB-D Image Analysis and Processing which was published by Springer in 2019 as part of their Advances in Computer Vision and Pattern Recognition book series.
BMVC2019

British Machine Vision Conference

I was on the organising committee for the 2019 British Machine Vision Conference which was held in Cardiff.
IVF2019

Image and Video Forensics

With Christian Riess (Friedrich-Alexander University), Anthony Hoogs (Kitware), and Xianfang Sun (Cardiff University) I organised a workshop on Image and Video Forensics which was held in conjunction with WACV 2019.

Non-Rigid Shape Correspondence of Isometric and Non-Isometric Deformations

I co-organised the SHREC'19 competition Non-Rigid Shape Correspondence of Isometric and Non-Isometric Deformations which was held at the Workshop on 3D Object Retrieval (3DOR) which is collocated with Eurographics 2019.
scroll

Virtual unrolling of scrolls

Our work on the virtual unrolling of scrolls was featured on the BBC Radio Wales - Science Cafe and the One Show on BBC One . Also check out the University's press release, an earlier press release , and my webpage for more details.
RGBD18

RGB-D - sensing and understanding via combined colour and depth

With Yukun Lai (Cardiff University) and Toby Breckon (Durham University) I organised a workshop on RGB-D - sensing and understanding via combined colour and depth which was held in in conjunction with ACCV 2018.
VG2017

Vision meets Graphics

With Taehyun Rhee and Fang-Lue Zhang (Victoria University of Wellington) and Yukun Lai (Cardiff University) I organised a workshop on Vision meets Graphics which was held in in conjunction with PSIVT 2017.
RGBD

Analysis and Processing of RGBD Data

I co-organised with Yukan Lai a BMVA Technical meeting on Analysis and Processing of RGBD Data held on 1st March 2017.

NPRportrait (version 0.1)

The 40 images that make up the NPRportrait benchmark (version 0.1) described by David Mould (Carleton University) and myself at Expressive 2017 are available.

NPRgeneral

The 20 images that make up the NPRgeneral benchmark described by David Mould (Carleton University) and myself at Expressive 2016 are available.
VG2015

Vision meets Graphics

With Taehyun Rhee (Victoria University of Wellington) I organised a workshop on Vision meets Graphics held in Auckland on 23rd November 2015 in conjunction with PSIVT 2015.
CAe

Expressive Symposium

I co-organised the International Symposium on Computational Aesthetics in Graphics, Visualization, and Imaging (Expressive 2015).

Canonical Forms

We organised the SHREC'15 competition Canonical Forms for Non-Rigid 3D Shape Retrieval . The results were presented at the Workshop on 3D Object Retrieval (3DOR) which was collocated with Eurographics 2015.
cover

Cellular Automata in Image Processing and Geometry

Along with Andy Adamatzky and Xianfang Sun I edited a book on Cellular Automata in Image Processing and Geometry which was published by Springer in 2014 as part of their Emergence, Complexity and Computation book series.
CAe

Computational Aesthetics

I co-organised the Int. Symp. on Computational Aesthetics in Graphics, Visualization, and Imaging which was part of Expressive 2014, and collocated with SIGGRAPH.
NPRcover

Image and Video-Based Artistic Stylisation

Along with John Collomosse I edited a book on Image and Video-Based Artistic Stylisation which was published by Springer in 2013 as part of their Computational Imaging and Vision book series.
PAESNPR13

Photographic Aesthetics and Non-Photorealistic Rendering

With Ramakrishna Kakarala (Nanyang Technological University) I organised the PAESNPR13 workshop held in conjunction with APSIVT 2013.
ACCV logo

Face analysis: The intersection of computer vision and human perception

With colleagues I organised a workshop in conjunction with ACCV 2012.
PR logo

Agent based Computer Vision

I was co-guest editor of the 2004 special issue of Pattern Recognition on Agent based Computer Vision. This was the original call for papers. Here is its table of contents.
BMVC logo

British Machine Vision Conference

The British Machine Vision Conference (BMVC2002) was held in Cardiff in 2002! You can read about some of the work described at the conference in an article in HERO
JInt logo

Neural-Fuzzy Applications in Computer Vision

I was co-guest editor of the 2000 special issue of the Journal of Intelligent and Robotic Systems on Neural-Fuzzy Applications in Computer Vision. This was the editorial, setting the scene. Here is its table of contents.
JInt logo

Machine Vision using Neural Networks

I was co-guest editor of the 1998 special issue of Neural Computing and Applications on Machine Vision using Neural Networks. Here is its table of contents.

I have organised some of the BMVA technical meetings. You can read the abstracts for Low Level Feature Extraction, AI Methods and Data Fusion in Remote Sensing Image Processing and Analysis and Processing of RGBD Data.

My Erdős number is 3 through: me → Žunić → Ivić → Erdős


Software

Some software that I have developed (some of it in collaboration with colleagues) is available for: See also software by colleagues for our joint work:

Demos

You can upload your images and try out our non-photorealistic rendering method for generating line drawings.
You can upload your images and try out our APDrawingGAN neural style transfer of portraits as line drawings.

Data

You can download the images used in the paper "An alternative approach to computing shape orientation with an application to compound shapes", Jovisa Zunic and Paul L. Rosin, International Journal of Computer Vision, vol. 81, no. 2, pp. 138-154, 2009.
The annotations and extracted audio-visual features from the sequences of the two way conversations making up the Cardiff Conversational Database (CCDb) are available.
SHREC21 The data we provided for the SHREC'21 competition Quantifying Shape Complexity is available.
The data we provided for the SHREC'20 competition Shape Correspondence with Non-Isometric Deformations is available.
The data we provided for the SHREC'20 competition Non-Rigid Shape Correspondence of Physically Based Deformations is available.
The data we provided for the SHREC'19 competition Non-Rigid Shape Correspondence of Isometric and Non-Isometric Deformations is available.
The data we provided for the SHREC'15 competition Canonical Forms for Non-Rigid 3D Shape Retrieval is available.
SHREC The data we provided for the SHREC'14 competition Shape Retrieval of Non-Rigid 3D Human Models is available.
The 60 images that make up the NPRportrait benchmark (version 1.0) are available.
The 20 images that make up the NPRportrait benchmark described by David Mould (Carleton University) and myself at Expressive 2017 are available.
The artist line drawings and original images that provided training data for APDrawingGAN are available.
The 20 images that make up the NPRgeneral benchmark described by David Mould (Carleton University) and myself at Expressive 2016 are available.
solder The solder joint image dataset from our compressive sensing paper is available here.
The FRITH dataset of tampered images is available.
Multi-institution microscope slides and herbarium sheets with ground truth for semantic and instance segmentation.
The artistic image aesthetics dataset BAID is available.

Publications

You can see a list of my papers with many PDF files available too:

Contact Details

Address:
Room 2.63
School of Computer Science & Informatics
Cardiff University
Abacws
Senghennydd Road
Cardiff
CF24 4AG
United Kingdom
 
Telephone Number:
+44 (0)29 2087 5585
 
Fax Number:
+44 (0)29 2087 4598
 
Email:
Paul.Rosin@cs.cf.ac.uk