Annual all-hands meetings:
The annual V&L Net Meetings will be a combination of research conference and networking event. We want these events to be exciting, innovative and high-profile, showcasing the best of what V&L Net does. We will invite keynote speakers who are research leaders in V&L Net topics, and organise sessions of short oral presentations, poster sessions and demo sessions. We will invite industry representatives, seek industrial sponsorship, and organise smaller technical meetings on specific topics.
In terms of networking, we will employ a diverse range of networking techniques, such as working lunches, speed-dating sessions and `dance cards'. Coffee breaks, poster sessions and an art exhibition (see below) will provide further opportunities for networking.
The annual V&L meetings will also have small, professionally curated exhibitions of generative/interactive art that combines language and vision. These will serve the dual purpose of (i) bringing together scientific researchers and artists working on similar topics, but from very different perspectives which we expect to stimulate new ideas on both sides; and (ii) increasing the general interest of the meetings to the CV and NLP communities, as well as the media.
We will organise (and encourage other network members to organise) research workshops co-located with leading international conferences in the vision and language fields. These workshops will have strong identifying themes, and will be aimed at drawing in members of the wider vision and language processing communities. The workshops will be sponsored by V&L Net (financial support limited to invited speaker costs, as parent conferences will charge workshop fees), but independently organised, and will publish proceedings. To encourage interest and involvement we will aim to publish extended versions of selected workshop papers in collections or special issues in relevant journals such as in IET Computer Vision and Computational Linguistics.
Examples workshops are: a Workshop on Text-to-Image Generation at the Annual Meeting of the Association for Computational Linguistics (ACL); a workshop on Automatic Tagging and Annotation of Visual Content at the International Conference on Computer Vision (ICCV); a one-day technical meeting on Bridging Between Visual Semantics and Language Semantics co-organised with the British Machine Vision Association (BMVA). In order to counteract a bias in attendance at such workshops between CV and NLP researchers resulting from the event being co-located with either a CV or NLP conference, we will invite speakers and panel members from the potentially under-represented discipline, and advertise the workshops extensively to mailing lists, inviting submissions.