self driving car CNN powerVRIf you’re tired of all the noise around the U.S. election the initials CNN are probably not what you want to be seeing right now. However, in our upcoming webinar CNNs are exactly what we’ll be talking about. However, in this case, by CNN we are referring to convolutional neural networks, which are set to be a key tool in enabling the self-driving cars of the future.

In their basic form CNNs are a form of neural networks used to train computers to recognise objects in a similar fashion to how the human brain does so. A neural network uses machine learning algorithms to take input signals, such as a picture of a cat, a car or a person, and turn them into output signals, which refers to the successful labelling of that input.

The more information you feed a CNN the better chance it has of succesfully recognising an object – much like a person in fact. This involves the use of deep learning techniques to create advanced AI. Yes, we’re talking deep, deep learning. These techniques will ensure that self-driving cars have a greater sense of the world around them, which will accelerate their use and acceptance in society at large.

In our upcoming webinar, Bryce Johnstone, Director of Automotive Segment Marketing at Imagination Technologies, together with Paul Brasnett, Principle Research Engineer, Imagination Technologies will be talking about how improving and speeding up image recognition is going to be key for creating the advanced self-driving cars of the future.

They will also explain how the GPU Compute power of Imagination’s PowerVR graphics chips make then ideally suited to to the task of running CNNs compared to traditional CPUs.

You can register for this webinar online and if you miss it live it will also be available to view on-demand.

Overview

Title: Leveraging PowerVR GPU Compute for Automotive CNN

Date: Wednesday, November 09, 2016

Time: 11:00 AM Pacific Standard Time

Duration: 30 minutes

 

webinar CNNs automotive powerVR

Comments