Connected and Autonomous Vehicles: Computer Vision and AI Unit 4 - Deep Learning for Machine Vision
Deep learning is a large field, that expands beyond the scope of computer vision, as you will have observed in the previous lesson plan.
However, in this course, we focus on the context of computer vision. Specifically, there are a number of deep learning approaches that may be relevant to autonomous vehicles.
Namely, region proposal networks, and segmentation networks are often used within this domain.
These are machine learning algorithms that have been trained to detect objects within an image, and output a bounding box around that object, or to state which pixels of the image belong to a specific object within the scene, providing a finer grain of detail than that of simply a bounding box.
However, these networks first rely on the concept of a Convolutional Neural Network, so first we need to cover the basics of CNN’s
All prices shown are exclusive of VAT which will be added at checkout.
|Partner Details||The Transport Systems Catapult (TSC) is the UK’s technology and innovation centre for Intelligent Mobility. It is a neutral, not-for-profit, Technology and Innovation company which aims to make the UK a world leader in Transport Innovation.|
• Identify key reasons for the use of human computation.
• Give a high-level overview of Segnet.