Swin Transformer
Swin Transformer is a hierarchical vision Transformer that employs local window attention and a shifted window mechanism, reducing attention complexity from quadratic to linear.
Timeline (last 90 days)
The genealogy of computer vision models is more miscellaneous than that of language models, including LeNet, AlexNet, VGG, ResNet, DenseNet, EfficientNet, Vision Transformer, Swin, ConvNeXt, and others, with each architecture solving a specific defect exposed by the previous architecture.
SwinIR introduced the Swin Transformer architecture to better capture long-range pixel dependencies.
Swin Transformer introduced a local window attention mechanism.
All Facts (3)
The genealogy of computer vision models is more miscellaneous than that of language models, including LeNet, AlexNet, VGG, ResNet, DenseNet, EfficientNet, Vision Transformer, Swin, ConvNeXt, and others, with each architecture solving a specific defect exposed by the previous architecture.
50%UnverifiedSwinIR introduced the Swin Transformer architecture to better capture long-range pixel dependencies.
50%UnverifiedSwin Transformer introduced a local window attention mechanism.
50%