Past conferences: 2014 | 2015
  • Русский
  • English

Ihor Starepravo

Director, Embedded at Luxoft

Ihor Starepravo

Director, Embedded at Luxoft


Over 20 years experience in the Software development field, including multimedia, video processing, machine vision and generic AI.


Topic: Embedded Vision Cases

Short Description: Although the foundation of Convolutional Neural and Deep Learning Neural Networks (CNN / DNN) around academic researches since 80’s, only recently the computing power emerged to mobile and server platforms enabled practical implementation of machine learning, pattern recognition and big-data analysis. Luxoft, together with its broad partners’ network, including companies as ARM and Imagination Technologies, is one of the world leaders in the development of CNN/DNN applications for real-time video processing, automotive applications. Alongside with server platforms we engage neural network technologies with embedded platforms that are traditionally limited with computing power as well as battery life. In my report, I will elaborate why artificial intelligence application development challenges the traditional paradigm of software programming. I will show you on the examples of our projects how does it work in today’s application and in which areas the use of neural networks could provide a new pace of information systems technological evolution.

All session by Ihor Starepravo

Embedded Vision Cases
Level: middle

09 Oct 2016
Stream #1

Presentation: Embedded Vision Cases
Level: middle