Real-Time Image Recognition Using MemSQL
The future of computing is visual. With everything from smartphone to Spectacles, we are about to see more digital imagery and associated processing than ever before.
In conjuction, new computing models are rapidly appearing to help data engineers harness the power of this imagery.
We will explore algebraic operations on tensors and examine how that can assist in large scale, high-throughput, highly parallel image recognition.
This presentation will feature a combinations of
_Architectural considerations in building and image recognition pipeline
_Advantages and pitfalls of specific approaches
_Real-time capabilities for instant matches
This talk features a live demonstration of constructing and executing a real-time image recognition pipeline.
Nikita Shamgunov co-founded MemSQL and has served as CTO since inception. Prior to co-founding the company, Nikita worked on core infrastructure systems at Facebook. He served as a senior database engineer at Microsoft SQL Server for more than half a decade. Nikita holds a bachelor’s, master’s and doctorate in computer science, has been awarded several patents and was a world medalist in ACM programming contests.