Head of Data Science Office, Eleks
Machine learning and artificial intelligence enthusiast, lecturer at Ukrainian Catholic University
ELEKS, Junior Data Scientist
During my work as a data scientist at ELEKS I participated in several projects related to computer vision and using Tensorflow as the main tool. I think the practical approaches we figured out while this work may be useful to everyone interested in deep learning and computer vision.
Тема доповіді: Deep learning, Tensorflow, and Fashion: how to stay in trend.
Recently artificial intelligence and computer vision has reached new performance standards and now plays a significant role in industries seemingly far from technology, like fashion. Our data science team faced such problem – clothes classification from surveillance cameras video data. We have used deep learning that is the most advanced and successful approach to such tasks and Tensorflow – a free and open source framework for numeric computations. I would like to share some practical tips about both contemporary state-of-the-art deep neural network architectures and implementations in Tensorflow.