deformable attention

可變形注意

在 Pytorch 中實現本文論文中的可變形注意力，這似乎是對 DETR 中提出的內容的改進。使用 SwinV2 中提出的連續位置嵌入，也對相對位置嵌入進行了修改，以實現更好的外推。

安裝

$ pip install deformable-attention

用法

 import torch
from deformable_attention import DeformableAttention

attn = DeformableAttention (
    dim = 512 ,                   # feature dimensions
    dim_head = 64 ,               # dimension per head
    heads = 8 ,                   # attention heads
    dropout = 0. ,                # dropout
    downsample_factor = 4 ,       # downsample factor (r in paper)
    offset_scale = 4 ,            # scale of offset, maximum offset
    offset_groups = None ,        # number of offset groups, should be multiple of heads
    offset_kernel_size = 6 ,      # offset kernel size
)

x = torch . randn ( 1 , 512 , 64 , 64 )
attn ( x ) # (1, 512, 64, 64)

3d 可變形注意力

 import torch
from deformable_attention import DeformableAttention3D

attn = DeformableAttention3D (
    dim = 512 ,                          # feature dimensions
    dim_head = 64 ,                      # dimension per head
    heads = 8 ,                          # attention heads
    dropout = 0. ,                       # dropout
    downsample_factor = ( 2 , 8 , 8 ),      # downsample factor (r in paper)
    offset_scale = ( 2 , 8 , 8 ),           # scale of offset, maximum offset
    offset_kernel_size = ( 4 , 10 , 10 ),   # offset kernel size
)

x = torch . randn ( 1 , 512 , 10 , 32 , 32 ) # (batch, dimension, frames, height, width)
attn ( x ) # (1, 512, 10, 32, 32)

1d 可變形注意力以達到良好的測量效果

 import torch
from deformable_attention import DeformableAttention1D

attn = DeformableAttention1D (
    dim = 128 ,
    downsample_factor = 4 ,
    offset_scale = 2 ,
    offset_kernel_size = 6
)

x = torch . randn ( 1 , 128 , 512 )
attn ( x ) # (1, 128, 512) 

引文

 @misc { xia2022vision ,
    title   = { Vision Transformer with Deformable Attention } , 
    author  = { Zhuofan Xia and Xuran Pan and Shiji Song and Li Erran Li and Gao Huang } ,
    year    = { 2022 } ,
    eprint  = { 2201.00520 } ,
    archivePrefix = { arXiv } ,
    primaryClass = { cs.CV }
}

 @misc { liu2021swin ,
    title   = { Swin Transformer V2: Scaling Up Capacity and Resolution } ,
    author  = { Ze Liu and Han Hu and Yutong Lin and Zhuliang Yao and Zhenda Xie and Yixuan Wei and Jia Ning and Yue Cao and Zheng Zhang and Li Dong and Furu Wei and Baining Guo } ,
    year    = { 2021 } ,
    eprint  = { 2111.09883 } ,
    archivePrefix = { arXiv } ,
    primaryClass = { cs.CV }
}