Deep Learning / 23
Vision Transformers
Explore image patches, embeddings, vision attention, and comparisons between CNNs and ViTs.
This is a knowledge index. Published titles open their articles; the remaining entries outline planned coverage.
Subtopics
- 23.01
Why Vision Transformers?
Planned guide: Why Vision Transformers. Focus on intuition, assumptions, and a worked example.
- 23.02
Image Patches
Planned guide: Image Patches. Focus on intuition, assumptions, and a worked example.
- 23.03
Patch Embeddings
Planned guide: Patch Embeddings. Focus on intuition, assumptions, and a worked example.
- 23.04
Positional Embeddings
Planned guide: Positional Embeddings. Focus on intuition, assumptions, and a worked example.
- 23.05
Vision Self-Attention
Planned guide: Vision Self-Attention. Focus on intuition, assumptions, and a worked example.
- 23.06
CLS Token
Planned guide: CLS Token. Focus on intuition, assumptions, and a worked example.
- 23.07
ViT Architecture
Planned guide: ViT Architecture. Focus on intuition, assumptions, and a worked example.
- 23.08
CNN vs ViT
Planned guide: CNN vs ViT. Focus on intuition, assumptions, and a worked example.
- 23.09
Hybrid Vision Architectures
Planned guide: Hybrid Vision Architectures. Focus on intuition, assumptions, and a worked example.
- 23.10
Vision Transformer Scaling
Planned guide: Vision Transformer Scaling. Focus on intuition, assumptions, and a worked example.