Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

Stephen Lin, Zheng Zhang, Han Hu, Yue Cao, Baining Guo, Ze Liu, Yutong Lin, Yixuan Wei

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 207 functions out of this paper's own repositories and ran 120 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

Showing the first 60. An agent gets every one from get_harvested_code_for_paper("2103.14030").

Repositories linked to this paper

• microsoft/Swin-Transformer — marked official in Papers with Code; named in the paper
• rwightman/pytorch-image-models — named in the paper
• AntixK/PyTorch-Model-Compare
• BR-IDL/PaddleViT
• Burf/SwinTransformer-Tensorflow2
• DarshanDeshpande/jax-models
• DominickZhang/Distillation-Swin-Transformer
• Gojay001/toolkit-DeepLearning
• IMvision12/keras-vision-models
• JIA-HONG-CHU/Swin-Transformer-add-EncNet-DaNet-DraNet-for-semantic-segmentation-on-Statelite-Dataset
• LiWentomng/OrientedRepPoints
• Myyyr/transseg2d
• NEUdeep/Swin-Transformer-Object-Detection
• SforAiDl/vformer
• SwinTransformer/Transformer-SSL
• USTC-IMCL/HST-for-Compressed-SR
• VcampSoldiers/Swin-Transformer-Tensorflow
• WangFeng18/Swin-Transformer
• YongWookHa/swin-transformer-ocr
• abman23/pmnet
• ayanglab/swinganmr
• ayanglab/swinmr
• befallenStar/molecularAtt
• berniwal/swin-transformer-pytorch
• canerozer/qct
• facebookresearch/hiera
• fogfog2/packnet
• holdfire/CLS
• holdfire/FAS
• huggingface/transformers

More are linked than shown here.

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

This paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such as large variations in the scale of visual entities and the high resolution of pixels in images compared to words in text. To address these differences, we propose a hierarchical Transformer whose representation is computed with Shifted windows. The shifted windowing scheme brings greater efficiency by limiting self-attention computation to non-overlapping local windows while also allowing for cross-window connection. This hierarchical architecture has the flexibility to model at various scales and has linear computational complexity with respect to image size. These qualities of Swin Transformer make it compatible with a broad range of vision tasks, including image classification (87.3 top-1 accuracy on ImageNet-1K) and dense prediction tasks such as object detection (58.7 box AP and 51.1 mask AP on COCO testdev) and semantic segmentation (53.5 mIoU on ADE20K val). Its performance surpasses the previous state-of-theart by a large margin of +2.7 box AP and +2.6 mask AP on COCO, and +3.2 mIoU on ADE20K, demonstrating the potential of Transformer-based models as vision backbones. The hierarchical design and the shifted window approach also prove beneficial for all-MLP architectures. The code and models are publicly available at https://github. com/microsoft/Swin-Transformer.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2103.14030")
get_code_for_paper("2103.14030")
have("2103.14030")

Connect an agent — have() is free.