We are very excited to announce that our next DL meetup @ Dublin 2018 is coming soon and will be hosted at Zalando Dublin office on 20th of June from 6 pm to 9 pm.
The Zalando Fashion Insights Centre was founded with the aim of understanding fashion through technology. We’ve set ourselves the goal of organising the world’s fashion knowledge. https://jobs.zalando.com/tech/blog/dublin-tech-hub-three/
The topic is "Summer Events @ Zalando -- Deep Learning in Vision and Text". We have proudly invited 4 speakers to share their knowledge and experience on applying deep learning technologies to solve real-world problems such as image-tagging for Fashion content image, video analytics for dense crowd monitoring, and so on. We expect this event to be a high-quality and insightful event as always.
[18:00 - 18:30] Registration, Networking, Welcome Message by Zhenxing Zhang.
[18:30 - 18:50] Apply deep learning to work in practice -- Dr Robert Hogan -- Data Scientist, Zalando Fashion Insight Centre Dublin.
"Content will be coming up very soon"
[19:00- 19:20] Relation Extraction from Unstructured Texts and its Applications -- Dr. Jinhua Du -- Research Fellow at ADAPT Centre for Digital Content Technology
"In this talk, I will first give a brief introduction to relation extraction, and then introduce the related work from different points of view: (1) workflow; (2) methods; (3) data sets. Thirdly, I will talk about our ongoing work using deep neural networks and multi-instance learning framework. Experimental results on different data sets and result analysis will also be given. Finally, I will briefly introduce the application of relation extraction in Finance and other domains."
[19:30 - 19:50] Multi-Modal Methods: Visual Speech Recognition and Image Captioning -- Ben Duffy -- Data Scientist in Accenture and Co-founder/Lead Researcher at The M Tank
"I'll be talking about some recent intersections between Computer Vision (CV) and Natural Language Processing (NLP). Specifically advances within the fields of Visual Speech Recognition (AKA automated lip reading) and Image Captioning, which have largely been enabled by Deep Learning techniques e.g. plugging Convolutional Neural Networks (CNN) into Recurrent Neural Networks (RNN). The talk will be formatted similarly to the 3 part publication written by The M Tank which is hosted at http://www.themtank.org/multi-modal-methods"
[20:00 - 20:20] HiData -- Dr. Leonardo Gualano -- Post Dr in the Insight Centre for Data Analytics, DCU
"The project (HiData) is about video analytics for dense crowd monitoring (dense crowd counting, demographics, detection of groups within crowd...). We will talk about our work in progress in this specific project, what we have been worked on till now and what we aim to do in future."
[20:30 - 20:50] Networking and Closing.