MIR.models.Deformable_Swin_Transformer_v2

Deformable Swin Transformer v2 backbone for 3D registration.

Functions

deform_window_partition(x, window_size)
window_partition(x, window_size)
window_reverse(windows, window_size, H, W, L)

Classes

BasicLayer(*args, **kwargs)	A basic Swin Transformer layer for one stage.
DefSwinTransformerV2(*args, **kwargs)	Deformable Swin Transformer
LayerNormProxy(*args, **kwargs)	LayerNorm applied over channel-last representation.
Mlp(*args, **kwargs)	Feed-forward MLP block used inside transformer blocks.
Offset_block(*args, **kwargs)	Offset prediction block for deformable attention.
Offset_block0(*args, **kwargs)	Offset prediction block for deformable attention (lightweight).
PatchEmbed(*args, **kwargs)	Image to Patch Embedding :param patch_size: Patch token size.
PatchMerging(*args, **kwargs)	Patch Merging Layer.
SinPositionalEncoding3D(*args, **kwargs)
SinusoidalPositionEmbedding(*args, **kwargs)	Rotary Position Embedding
SwinTransformerBlock(*args, **kwargs)	Swin Transformer Block.
WindowAttention(*args, **kwargs)	Window based multi-head self attention (W-MSA) module with relative position bias.
XMlp(*args, **kwargs)