Statements (30)
Predicate | Object |
---|---|
gptkbp:instanceOf |
gptkb:model
vision transformer |
gptkbp:architecture |
hierarchical transformer
|
gptkbp:arXivID |
2103.14030
|
gptkbp:author |
Ze Liu
|
gptkbp:citation |
over 5000 (as of 2024)
|
gptkbp:designedFor |
computer vision
|
gptkbp:developedBy |
gptkb:Microsoft_Research_Asia
|
gptkbp:frameworkSupport |
gptkb:TensorFlow
gptkb:PyTorch |
https://www.w3.org/2000/01/rdf-schema#label |
Swin Transformer
|
gptkbp:improves |
ViT on several vision tasks
|
gptkbp:influenced |
subsequent vision transformer models
|
gptkbp:input |
gptkb:illustrator
|
gptkbp:introducedIn |
2021
|
gptkbp:notableFeature |
shifted window attention
|
gptkbp:notableFor |
remote sensing
autonomous driving medical image analysis |
gptkbp:notablePublication |
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
|
gptkbp:openSource |
yes
|
gptkbp:publicationDate |
gptkb:arXiv
|
gptkbp:repository |
https://github.com/microsoft/Swin-Transformer
|
gptkbp:usedFor |
image classification
object detection semantic segmentation |
gptkbp:usedIn |
ADE20K semantic segmentation
COCO object detection |
gptkbp:bfsParent |
gptkb:convolutional_neural_network
|
gptkbp:bfsLayer |
5
|