// ◉ Multimodal & Vision
Swin Transformer
Hierarchical vision Transformer with shifted local-window attention.
2021 Intermediate extended
IN PLAIN WORDS
ViT attends globally, so cost explodes at high resolution. Swin attends inside small windows, shifts the windows by half a window in the next layer so information flows across them, and merges patches layer by layer into multi-scale features like a CNN – making Transformers practical for detection and segmentation.
Used in: Backbone for detection, segmentation and video understanding
01Original paper
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Ze Liu, Yutong Lin, Yue Cao, et al. · 2021
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
Then you can learn
- Nothing in the atlas builds on it yet
Related:Sliding-Window & Sparse AttentionDETR (Detection Transformer)
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
04Where the course teaches it
For AI / data
- Verification: verified, checked 2026-10-22.
- JSON: /api/learn/algorithms.json (id:
swin)