Proceedings of the Machine Vision of Animals and their Behaviour Workshop (MVAB) 2015

Editors: T. Amaral, S. Matthews, T. Plötz, S. McKenna, and R. Fisher BMVA Press, September 2015, ISBN 1-901725-57-X


The Machine Vision of Animals and their Behaviour (MVAB) workshop brought together members of the community researching computer vision for animals, from such diverse application areas as wildlife study, animal farming, and industrial inspection.

We were pleased to present a programme featuring nine papers by authors from six countries (mostly in Europe, with a contribution from Australia), affiliated to 13 universities, two companies and one governmental body. Three papers were related with food production (milk, chicken livers, and fish), two of them addressing post-mortem inspection. The remaining six papers focused on the study of animals in the wild, ranging from very small brine shrimps to great white sharks, not forgetting crested black macaques, various birds, reef fish, and butterflies.

Echoing the diversity of application areas covered in the selected papers, the workshop’s oral sessions started with two stimulating keynote speeches from very different perspectives: Prof. Ilias Kyriazakis addressed aspects of livestock production that can potentially benefit from the use of computer vision, whereas Prof. Robert Fisher discussed the wide range of applications of the Fish4Knowledge database.

From the onset we envisaged that all accepted papers would be published, in order to make the call for papers as appealing as possible. Each paper was reviewed by three programme committee members in a double-blind process and, ultimately, we accepted all papers that had no recommendation for rejection from any reviewer. This resulted in nine accepted manuscripts out of 15 submissions, or a 60\ 0x1.dfbd44p-1015cceptance rate. In order to ensure time for discussion after each oral presentation, while keeping the talks themselves reasonable in length, we scheduled six oral presentations and dedicated a poster session to the remaining three papers.

A best paper certificate was awarded to Geoffrey French, Mark H. Fisher, Michal Mackiewicz, and Coby L. Needle, from the University of East Anglia and Marine Scotland, for their paper on “Convolutional Neural Networks for Counting Fish in Fisheries Surveillance Video”, which received the highest reviewer scores.

We express our gratitude to all authors who submitted their work, making the workshop possible; to the members of the programme committee, for their timely reviewing work, which enabled us to come up with a balanced, high-quality programme; to Prof. Ilias Kyriazakis, for his availability to contribute with a challenging keynote speech; to the BMVC organisers, Xianghua Xie, Gary Tam, and Mark Jones, for their prompt help with all our administrative queries; and naturally to all delegates who chose to participate in our workshop. Our thanks also for the permission to use some images from the papers to illustrate the workshop's programme and web page, granted by the authors and by Michael C. Scholl of the Save Our Seas Foundation (shark fin image).

The MVAB chairs,
Telmo Amaral, Stephen Matthews, Thomas Plötz, Stephen McKenna, and Robert Fisher

Programme Committee

Alexandra Branzan Albu (University of Victoria, Canada)
Douglas Armstrong (University of Edinburgh, UK)
Vinay Bettadapura (Georgia Institute of Technology, US)
Elsbeth van Dam (Noldus Information Technology, Netherlands)
Patrick Dickinson (University of Lincoln, UK)
Alfonso Pérez-Escudero (Massachusetts Institute of Technology, US)
Yasuyo Kita (National Institute of Advanced Industrial Science and Technology, Japan)
Anders Ringgaard Kristensen (University of Copenhagen, Denmark)
Eric Pauwels (Centrum Wiskunde & Informatica, Netherlands)
Sai Ravela (Massachusetts Institute of Technology, US)
Georgios Tzimiropoulos (Nottingham University, UK)