Workshops

BMVC 2019 features 5 workshops, detailed below. Workshops will take place on the afternoon of Thursday 12th September 2019.

For any general inquiries about workshops at BMVC 2019, please contact the workshop chairs using workshops@bmvc2019.org. For inquiries about specific workshops, please use the contact details listed below.

Applications of Egocentric Vision Workshop (EgoApp)

Location: E0.09
Workshop Website: https://egoappworkshop.wordpress.com

EgoApp aims to bring together diverse communities related to egocentric vision, such as computer science or social science to discuss the current and next generation of related technologies. To this end, both research and industry communities are invited to submit their recent outcomes in the format of a paper or a position paper to EgoApp. Submissions are expected to deal with egocentric vision including, but not limited to:

  • Wearable technologies for egocentric vision acquisition, computation, and perception
  • Assistive technologies for personalized monitoring
  • Object detection and recognition from egocentric vision
  • Person identification in egocentric videos
  • Activity recognition in egocentric vision
  • Visual lifelogging and summarization of egocentric videos
  • Scene understanding from egocentric vision
  • Human computer/robot interaction using egocentric vision
  • Social signal analysis and behaviour modelling in egocentric vision
  • Understanding of group dynamics in egocentric vision
  • Attention, fixation, and saliency modelling and prediction
  • Perception in virtual and augmented reality
  • Ethics and social issues in egocentric vision

Maya Aghaei 
Postdoctoral Researcher 
Pattern Analysis and Computer Vision, Istituto Italiano di Tecnologia, Italy maya.aghaei@iit.it 

Cigdem Beyan 
Postdoctoral Researcher 
Pattern Analysis and Computer Vision, Istituto Italiano di Tecnologia, Italy cigdem.beyan@iit.it 

Fernando De la Torre
Research Scientist Manager
Facebook Research, USA ftorre@andrew.cmu.edu  

Vittorio Murino
Principal Investigator 
Pattern Analysis and Computer Vision, Istituto Italiano di Tecnologia, Italy vittorio.murino@iit.it

 Embedded AI for Real-Time Machine Vision

Location: C-1.04
Workshop Website: https://sites.google.com/view/emai2019/home

Real-time machine vision based on AI and ML approaches is a critical technology for autonomous vehicles, smart cities, and industrial computer vision. Advances in 3D VLSI allow us to integrate sensing, computation, and memory in a single platform that provides much higher performance and lower power than is possible even with traditional embedded computer vision platforms. AI and machine learning methods can be tightly integrated with sensing to create small, low-power, high-performance embedded machine vision systems. However, the realization of these systems will require co-design with innovations for both algorithms and architectures. Potential topics include:

  • Analog and mixed-signal on-sensor computation.
  • Digital on-sensor computation.
  • 3D VLSI architectures for embedded AI.
  • Computer vision applications (tracking, target identification, etc.) adapted to a chosen real-time AI vision platform.
  • Algorithm-architecture co-design for embedded AI vision.
  • Simulation studies of real-time embedded AI vision systems.
  • Benchmarks for evaluation of embedded AI vision systems

Saibal Mukhopodhyay
Georgia Tech, saibal@ece.gatech.edu  

Marilyn Wolf
Georgia Tech, wolf@ece.gatech.edu 

Interpretable and Explainable Machine Vision

Location: E1.21
Workshop Website: https://sites.google.com/view/ixmv2019/

Recent years have seen significant advances in techniques for image processing and machine vision based on breakthroughs in machine learning and artificial intelligence, especially in the area of deep neural networks. However, such techniques are widely viewed as creating “blackbox” systems that are in some senses “inscrutable”, leading to concerns over their reliability,stability, and trustworthiness. Consequently, we have seen a surge of interest in approaches aimed at “opening the black boxes” commonly characterised by the terms ​interpretability ​and explainability. Topics of interest include (but are not limited to):

  • Improving the theoretical basis of interpretability and explainability techniques
  • Practical interpretability and explainability techniques for system developers
  • Case studies of applied interpretability and explainability approaches
  • Visualisation techniques for network layer representations
  • Evaluation metrics for interpretability
  • Psychological and human-in-the-loop perspectives on interpretability and explainability
  • Approaches aimed at assuring fairness, accountability and transparency
  • Interpretable model architectures
  • Explaining and interpreting uncertain

Alun Preece
PreeceAD@cardiff.ac.uk
Cardiff University, UK

Supriyo Chakraborty
IBM Research, USA

Lewis Griffin
UCL

Mark Hall
Airbus, UK

Simon Julier 
UCL

Richard Tomsett 
IBM Research, UK

Sriram Varadarajan
BAE Systems, UK

Chris Willis
BAE Systems, UK

Object Detection and Recognition for Security Screening

Location: C-1.01
Workshop Website: https://odrss.blogs.lincoln.ac.uk/

Security in transportation systems and public places is recognised as a global research priority: “Secure Societies” is a priority theme in Horizon 2020’s Societal Challenges, and the current work programme includes a number of calls for proposals during 2019-2020. The UK government is also undertaking a similar programmes of research funding, such as the Future Aviation Security Solutions Programme (FASS), committing £25m over 3 years to develop technological solutions to airport security. This workshop will bring together researchers working on (or interested in) the use of image analysis, computer vision and machine learning techniques for the detection and identification of threats, dangerous objects and other relevant objects of interest,for the purpose of security screening.This includes, but is not limited to, analysis of image capture in the X-Ray spectrum. The workshop will enable participants to present and share current work in this area, share and exchange ideas for future innovation, and build potential partnerships for future work, funding applications, and commercialisation with academic and industry partners. Topics of interest include but are not limited to:

  • Airport and Transportation Security
  • Public security screening
  • X-Ray Image Analysis
  • Machine Learning/Deep Learning
  • Object detection and Recognition
  • Autonomous Vehicle Platforms
  • Robust testing and deployment in the field

Professor John C. Murray
john.murray@hull.ac.uk
Professor of Robotics and Autonomous Systems
University of Hull, UK

Dr Patrick Dickinson
Principal Lecturer
University of Lincoln, UK 

Dr Tryphon Lambrou
Senior Lecturer
University of Lincoln, UK  

Professor Jason Ralph
Professor of Electrical Engineering
University of Liverpool, UK

Professor Toby Breckon
Professor of Computer Science
Durham University, UK 

VAIE2019: Workshop on Visual Artificial Intelligence and Entrepreneurship

Location: C0.07
Workshop Website: https://dculibrk.github.io/bmvc-vaie/

Global spending on AI technology is expected to reach $57 billion by 2021. An increase of 4.5 times over the funds invested in 2017, as diverse industries are investing aggressively in projects that utilize cognitive/AI software capabilities. Machine vision (visual AI) is expected to account for at least a quarter of that spending. The 1st Workshop on Visual AI and Entrepreneurship (VAIE) of the British Machine Vision Conference (BMVC 2019) aims to bring together researchers and industry practitioners in the domain of visual AI (computer vision) and more broadly AI and provide a forum to present and discuss research done with, for and by the industry, with preference given to the efforts focused on startups, SMEs and innovation activities within larger enterprises. As such, the topics of interest include, but are not limited to:

  • Visual Artificial Intelligence research done by startups, SMEs and industry
  • Industrial case studies and position papers focusing on applications of visual AI within startups, SMEs and industrial systems.
  • Demos of systems and prototypes based on visual artificial intelligence.
  • Statistics and machine learning for vision and multimedia processing.
  • Research studies on all the topics of interest to the BMVC itself, provided that a clear link between the research study and a startup, SME or industrial entity can be demonstrated in the camera-ready paper.

In addition to the classical BMVC paper format we plan to solicit submissions in the form of a short position paper or industrial case study that could be as short as three pages without references.

Prof. Dubravko Culibrk 
dubravko.culibrk@tandemlaunch.com  
Senior Scientist, Tandemlaunch Inc., Canada Full Professor, University of Novi Sad, Serbia 

Prof. Graham Taylor  
gwtaylor@uoguelph.ca  
Associate Professor and Canada Research Chair in Machine Learning
University of Guelph, Canada Academic Director, NextAI, Canada 

Prof. Niculae Sebe 
niculae.sebe@unitn.it 
Full Professor, Dept. of Information Engineering and Computer Science, University of Trento, Italy Research Expert, Huawei Technologies, Ireland