From the course: Hands-On Introduction to Transformers for Computer Vision
Join today to access over 24,900 courses taught by industry experts.
Why transformers are revolutionary for computer vision - PyTorch Tutorial
From the course: Hands-On Introduction to Transformers for Computer Vision
Why transformers are revolutionary for computer vision
- [Instructor] Welcome to chapter two, module three, why transformers are revolutionary for computer vision. We're going to take a sharp look here at exactly how transformers impacted the computer vision field and kind of roll in the last two modules into its logical conclusion here. So why are transformers revolutionary? We mentioned this paper, "Attention Is All You Need." Later on in the course, we're going to do a real deep dive into this paper and understand exactly all the different parts of it. But for right now, we're going to look at just the high level impact in computer vision. We mentioned the wall in the last module about how these pre-transformer issues kept coming up, whether on the language side it was long sequence memory or slow training, or on the vision side, it was these local receptive fields and rigid architectures. Transformers revolutionized ML for two main reasons. The first one is we finally had that context. We could look at the whole sentence and…
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.