Taras LishchenkoComputer Vision Engineer at DatAI

<< Back

Taras
Lishchenko

Computer Vision Engineer
at DatAI

Biography

First journey in Computer Vision started in startup based on face recognition for surveillance systems. Despite the fact that it failed in 2 years, I’ve got a lot of experience and till now I can’t imagine myself working in other field. With the understanding that AI still is not applicable to most of real-world problems I’m working with a great team which is doing best to make some advances in Object Detection, Face Recognition and make it helpful in retail.

Topic

Deep learning in Scale with Python and Gstreamer.

Video Surveillance is being popular for decades, but with recent advances in Deep Learning field there emerged a possibility to automate Video Surveillance and make it even smarter. Some of well-known companies (ex: Samsung, Nvidia) have released own open-source projects (DeepStream SDK, NNStreamer) that brings Deep Learning closer to Video Processing.
During the session we are going to build a simple prototype of automated video surveillance system in Python using experience of existing solutions based on Gstreamer. We are going to make it simple enough so every can benefit from fast prototyping and good enough to make it applicable to real-world cases.

back to top