skip to content

 

 

Tuesday 9th June 2015 to Friday 12th June 2015

Background

 
The processing and analysis of time-varying data (such as in videos), encompasses many challenges, such as how to manage time varying aspects in order to detect objects, track and analyse their behaviours so that what is happening in a sequence of images can be better understood.    For instance, the parsing of a video sequence to provide a narrative of what is occurring remains a challenge, in terms of who or what is involved and their interactions. CCTV for instance, is increasingly commonplace and has proved to be an invaluable tool in the investigation of crime ranging from petty theft to the global threat of terrorism. Dynamic imaging in biomedicine is used to investigate the functionality of organs down to the dynamics of cells and molecules. 3D conversion in media industry requires the processing of high definition videos.  But the ability to rapidly interpret and extract meaningful conceptual information from large data streams is still beyond the current state of the art. There is therefore a need to identify and gain a better understanding of the mathematical/technical challenges that need to be overcome to ensure progress in the field.
 

Aims and Objectives

 With support from BAE Systems, this workshop sought to investigate the area of the analysis of very large and complex data streams, such as for videos, where there is a need to improve the current state-of-the art techniques. Within this context, there is a need to understand what is currently possible and what the main technical challenges are. Three industry challenges will be presented (one on each day), covering the areas of security, medical imaging, Media and Creative industries. Including
The workshop brought together industrial and academic experts from a diverse set of backgrounds in mathematics, computer science and information engineering including:
 

  • Mathematical Analysis
  • Computer Vision
  • Probabilistic Modelling/Data Fusion
  • Video Processing/Machine Learning
  • Statistics/Probabilistic Modelling

 
A key aim was to communicate the state of the art and understand the mathematical challenges that need to be overcome to ensure progress in this field.   The output from these workshops will also feed into the forthcoming Isaac Newton Institute Programme on Variational methods and effective algorithms for imaging and vision which will be taking place between 29 August and 20 December 2017. The day will include lots of opportunities for interaction and networking with break-out and discussion sessions.

Registration and Venue

Registration was open to researchers and industry representatives from relevant/related fields. Financial support for accommodation is available to academics.

The workshop took place at the Isaac Newton Institute for Mathematical Sciences in Cambridge. Please see the Isaac Newton Institute A-Z for further information about the venue.