Deep Learning ADAS Breakthrough Demo at CES

Paul Myles

TU-Editor

---

2nd January 2020

A deep learning model is claiming a 25% uplift in ADAS accuracy ahead of its demonstration at 2020 Consumer Electronics Show (CES).

Automotive product engineering and lifecycle services company, QuEST Global, claims its system improves the accuracy in detection of traffic signs, pedestrians and traffic. It employs synthetic data representing various environmental conditions and terrains while deep learning enhances the performance of applications to improve interpretation of both simulated and real time data.

Berthold Puchta, global industry leader – transportation, at QuEST Global, said: “The automotive industry is rapidly adopting autonomous, connected, electric and shared technologies. However, lack of variety of data is one of the major hurdles faced by the industry. This deep learning model developed by our engineers has shown higher accuracy rates and it can continuously evolve while training of the deep learning models continue.”

— Paul Myles is a seasoned automotive journalist based in London. Follow him on Twitter @Paulmyles_