Visual Computing Research Day, 29th June 2018

Talbot Room 909.

1130-1230 Smaller visual computing group (staff) – discuss scientific groups and grant support (with regards to recent research away day)

1230-1330 Lunch (larger group also including PGs, RAs and staff)

1330-1430 Rynson W. H. Lau

1430-1445 Coffee


David George, Deep Learning Driven Active Framework for Segmentation of Large 3D Shape Collections

Mike Edwards, Deep Learning in Irregular Domains

Joss Whittle, Good architecture choices for generative models

1530 Tea/Discussion

Mike Edwards passes PhD viva

Many congratulations to Mike Edwards, who passed his PhD viva today. His thesis is titled “Representation Learning in Irregular Domains”. Prof. Reyer Zwiggelaar and Dr. Dima Damen are the external examiners. The viva is chaired by Dr. Adeline Paiement.

VAST Best Paper Award

Congratulations to Gary Tam and co-authors for their Best Paper Award at VAST 2016: An Analysis of Machine- and Human-Analytics in Classification, Gary K. L. Tam, Vivek Kothari, Min Chen,

machine-and-human-classification-analysisIn this work, we present a study that traces the technical and cognitive processes in two visual analytics applications to a common theoretic model of soft knowledge that may be added into a visual analytics process for constructing a decision-tree model. Both case studies involved the development of classification models based on the “bag of features” approach. Both compared a visual analytics approach using parallel coordinates with a machine-learning approach using information theory. Both found that the visual analytics approach had some advantages over the machine learning approach, especially when sparse datasets were used as the ground truth. We examine various possible factors that may have contributed to such advantages, and collect empirical evidence for supporting the observation and reasoning of these factors. We propose an information-theoretic model as a common theoretic basis to explain the phenomena exhibited in these two case studies. Together we provide interconnected empirical and theoretical evidence to support the usefulness of visual analytics.

PhD Studentship: Visualization Techniques Applied to Internet of Things Data

This project will apply advanced visualization and analysis research to the massive quantities of data produced by water internet connected sensors. Managing precious water reserves is a key challenge facing the world. The company we will work with is a world leading developer of sensors for the water industry producing products that measure water flow, pinpoint leaks, monitor flood risk from rivers, etc. All of these devices are connected to the internet allowing real-time monitoring of the environment and water distribution to a customer level. They are currently in a major infrastructure project to install leak, pressure and flow sensors throughout water distribution network.

Application deadline: Friday 25th November 2015. Start date: January 1st 2016. Further details and application.


3 year research posts (typically £27,864-£37,394)

Computer Science at Swansea has funding for 5 research posts. Get in touch with me if you would like to work in the area of Visual Computing. The deadline for application is likely to be December 4th 2015, and interested people should contact me well before that date since we need to write a competitive case. The criteria for selection are:

  • Achieved PhD before March 2013.
  • Outside UK, or not been resident in the UK for more than 12 months in last three years.

Seminar: Professor Takayuki Itoh, Ochanomizu University, Japan

Rectangle packing algorithm for tree and graph visualization
Robert Recorde, Monday July 27th at 3pm

Photo of Professor Takayuki Itoh

Abstract This talk introduces the speaker’s techniques on space-filling tree and network visualization. The former part of the talk introduces a tree visualization technique which represents hierarchy of the tree structure as nested rectangular regions. The latter part of the talk introduces a network visualization technique which clusters nodes according to their categories and connectivity, and then displays the clusters as rectangular regions. Both techniques are based on a rectangle packing algorithm which efficiently places the rectangular regions onto a display space. This talk also introduces several applications of the tree and network visualization techniques.

Biography Professor Takayuki Itoh is a full professor of the department of information sciences in Ochanomizu University, Japan since 2011. His research interest includes visualization, computer graphics, and multimedia. He is a vice director of the society for art and science in Japan, and a track chair of ACM Symposium on Applied Computing, Multimedia and Visualization track.
Takayuki Itoh has received his B.S., M.S., and Ph.D. degrees from Waseda University in 1990, 1992, and 1997, respectively. He has been a researcher at Tokyo Research Laboratory of IBM Japan during 1992 to 2005. He was also a visiting researcher at Carnegie Mellon University in 2000, and Kyoto University during 2003 to 2005. He has been an associate professor in Ochanomizu University since 2005, and a full professor since 2011.