Pilot AI Vision Framework: From Doorbells to Defense

Wednesday, May 23, 3:30 PM - 4:00 PM
Summit Track: 
Enabling Technologies
Exhibit Hall A-3

Pilot AI’s Vision Framework has enabled real-time detection, classification and tracking in thousands of devices, from consumer applications to federal contracts. Though diverse in end-user application, these use cases all share a common disadvantage: they are compute-constrained. Small consumer electronics are compute-constrained as BoM cost limits the amount of silicon that can be integrated, whereas federal use cases are compute-constrained since the problem (seeing from thousands of feet in the sky) requires processing a tremendous amount of data in real-time without reliable network connectivity. Scaling a single framework to enable the diverse set of hardware platforms these applications represent, from ultra-low power DSPs and microcontrollers to full-size GPUs, is what differentiates Pilot AI’s Vision Framework. We will introduce Pilot AI’s deep-learning based computer vision framework for compute-constrained devices and demonstrate this framework in real-world applications to motivate the drive towards embedded deep-learning.


Jonathan Su

CEO, Pilot AI

Jonathan Su co-founded Pilot AI and is the CEO. He received his PhD from Stanford CS, where he focused on computational math and high-performance computing. Prior to Pilot AI, Jon founded Phisix Fashion Labs, which specialized in high fidelity clothing simulation, which was acquired by eBay in 2014. Prior to Phisix, he also held positions at Intel Labs and Microsoft Research Asia. Working on deep learning computer vision algorithms, Pilot AI Labs is the world leader in edge-based perception.

See you at the Summit! May 20-23 in Santa Clara, California!
Register today and reserve your hotel room!