From the course: Synthetic Data: Advanced Concepts and Applications
Join today to access over 24,600 courses taught by industry experts.
Reducing the domain gap
From the course: Synthetic Data: Advanced Concepts and Applications
Reducing the domain gap
- Are there a lot of differences between your real and synthetic data? If so, the domain gap between your real and synthetic data might cause performance issues. After this lesson, you'll be able to describe some AI and signal processing approaches to reducing the domain gap. The first approach is domain adaptation. Domain refers the data distribution. Domain adaptation is the ability to apply an algorithm trained in one or more source domains to a different but related target domain. There are many techniques to align the synthetic, the source domain, to the real world, the target domain. For instance, in computer vision, synthetic data can simulate scenes, objects, and lighting conditions, but have difficulty replicating the complexities of real-world images. A common way around this problem is to do image style transfer through GAN based approaches, or through Fourier transform. They can be used to transfer the style…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.