【语义分割】2021-PVT2 CVMJ

论文题目：PVT v2: Improved Baselines with Pyramid Vision Transformer

论文链接: https://arxiv.org/abs/2106.13797

论文代码：https://github.com/whai362/PVT

论文翻译：PVT，PVTv2 - 简书 (jianshu.com)

1. 简介

计算机视觉中的Transformer最近取得了令人鼓舞的进展。在这项工作中，作者通过添加3个改进设计来改进原始金字塔视觉Transformer（PVTv1），其中包括：

具有卷积的局部连续特征；
具有zero paddings的位置编码，
具有平均汇集。

通过这些简单的修改，PVTv2在分类、检测和分割方面显著优于PVTv1。此外，PVTv2在ImageNet-1K预训练下取得了比近期作品（包括 Swin Transformer）更好的性能。

2. 网络

PVTv1[33]的主要局限性有以下三个方面:

(1)与ViT类似，在处理高分辨率输入时(如短边为800像素)，PVTv1的计算复杂度相对较大。

(2) PVTv1将一幅图像视为一组不重叠的patch序列，在一定程度上丧失了图像的局部连续性;

(3) PVTv1中的位置编码是固定大小的，对于处理任意大小的图像是不灵活的。这些问题限制了PVTv1在视觉任务中的性能。

2.1 总体架构

2.2 Linear Spatial Reduction Attention

用LinearSRA替代SRA。这里需要说明的一个问题，作者在PVTv1中说自己没用到卷积，但是在压缩K、V的时候使用的是Conv2D（参见github中代码）。在PVTv2中使用平均池化替代Conv2D。

2.3 Overlapping Patch Embedding(重叠补丁嵌入)

2.4 Convolutional FeedForward

3. 代码 pvt2-upernet

import torch
from torch import nn, Tensor
from torch.nn import functional as Fclass DropPath(nn.Module):"""Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).Copied from timmThis is the same as the DropConnect impl I created for EfficientNet, etc networks, however,the original name is misleading as 'Drop Connect' is a different form of dropout in a separate paper...See discussion: https://github.com/tensorflow/tpu/issues/494#issuecomment-532968956 ... I've opted forchanging the layer and argument names to 'drop path' rather than mix DropConnect as a layer name and use'survival rate' as the argument."""def __init__(self, p: float = None):super().__init__()self.p = pdef forward(self, x: Tensor) -> Tensor:if self.p == 0. or not self.training:return xkp = 1 - self.pshape = (x.shape[0],) + (1,) * (x.ndim - 1)random_tensor = kp + torch.rand(shape, dtype=x.dtype, device=x.device)random_tensor.floor_()  # binarizereturn x.div(kp) * random_tensorclass DWConv(nn.Module):def __init__(self, dim):super().__init__()self.dwconv = nn.Conv2d(dim, dim, 3, 1, 1, groups=dim)def forward(self, x: Tensor, H: int, W: int) -> Tensor:B, _, C = x.shapex = x.transpose(1, 2).view(B, C, H, W)x = self.dwconv(x)return x.flatten(2).transpose(1, 2)class MLP(nn.Module):def __init__(self, dim, hidden_dim, out_dim=None) -> None:super().__init__()out_dim = out_dim or dimself.fc1 = nn.Linear(dim, hidden_dim)self.fc2 = nn.Linear(hidden_dim, out_dim)self.dwconv = DWConv(hidden_dim)def forward(self, x: Tensor, H: int, W: int) -> Tensor:return self.fc2(F.gelu(self.dwconv(self.fc1(x), H, W)))class Attention(nn.Module):def __init__(self, dim, head, sr_ratio):super().__init__()self.head = headself.sr_ratio = sr_ratioself.scale = (dim // head) ** -0.5self.q = nn.Linear(dim, dim, bias=True)self.kv = nn.Linear(dim, dim * 2, bias=True)self.proj = nn.Linear(dim, dim)if sr_ratio > 1:self.sr = nn.Conv2d(dim, dim, sr_ratio, sr_ratio)self.norm = nn.LayerNorm(dim)def forward(self, x: Tensor, H, W) -> Tensor:B, N, C = x.shapeq = self.q(x).reshape(B, N, self.head, C // self.head).permute(0, 2, 1, 3)if self.sr_ratio > 1:x = x.permute(0, 2, 1).reshape(B, C, H, W)x = self.sr(x).reshape(B, C, -1).permute(0, 2, 1)x = self.norm(x)k, v = self.kv(x).reshape(B, -1, 2, self.head, C // self.head).permute(2, 0, 3, 1, 4)attn = (q @ k.transpose(-2, -1)) * self.scaleattn = attn.softmax(dim=-1)x = (attn @ v).transpose(1, 2).reshape(B, N, C)x = self.proj(x)return xclass Block(nn.Module):def __init__(self, dim, head, sr_ratio=1, mlp_ratio=4, dpr=0.):super().__init__()self.norm1 = nn.LayerNorm(dim)self.attn = Attention(dim, head, sr_ratio)self.drop_path = DropPath(dpr) if dpr > 0. else nn.Identity()self.norm2 = nn.LayerNorm(dim)self.mlp = MLP(dim, int(dim * mlp_ratio))def forward(self, x: Tensor, H, W) -> Tensor:x = x + self.drop_path(self.attn(self.norm1(x), H, W))x = x + self.drop_path(self.mlp(self.norm2(x), H, W))return xclass PatchEmbed(nn.Module):def __init__(self, c1=3, c2=64, patch_size=7, stride=4):super().__init__()self.proj = nn.Conv2d(c1, c2, patch_size, stride, patch_size // 2)self.norm = nn.LayerNorm(c2)def forward(self, x: Tensor) -> Tensor:x = self.proj(x)_, _, H, W = x.shapex = x.flatten(2).transpose(1, 2)x = self.norm(x)return x, H, Wpvtv2_settings = {'B1': [2, 2, 2, 2],  # depths'B2': [3, 4, 6, 3],'B3': [3, 4, 18, 3],'B4': [3, 8, 27, 3],'B5': [3, 6, 40, 3]
}class PVTv2(nn.Module):def __init__(self, model_name: str = 'B1') -> None:super().__init__()assert model_name in pvtv2_settings.keys(), f"PVTv2 model name should be in{list(pvtv2_settings.keys())}"depths = pvtv2_settings[model_name]embed_dims = [64, 128, 320, 512]drop_path_rate = 0.1self.embed_dims = embed_dims# patch_embedself.patch_embed1 = PatchEmbed(3, embed_dims[0], 7, 4)self.patch_embed2 = PatchEmbed(embed_dims[0], embed_dims[1], 3, 2)self.patch_embed3 = PatchEmbed(embed_dims[1], embed_dims[2], 3, 2)self.patch_embed4 = PatchEmbed(embed_dims[2], embed_dims[3], 3, 2)dpr = [x.item() for x in torch.linspace(0, drop_path_rate, sum(depths))]# transformer encodercur = 0self.block1 = nn.ModuleList([Block(embed_dims[0], 1, 8, 8, dpr[cur + i]) for i in range(depths[0])])self.norm1 = nn.LayerNorm(embed_dims[0])cur += depths[0]self.block2 = nn.ModuleList([Block(embed_dims[1], 2, 4, 8, dpr[cur + i]) for i in range(depths[1])])self.norm2 = nn.LayerNorm(embed_dims[1])cur += depths[1]self.block3 = nn.ModuleList([Block(embed_dims[2], 5, 2, 4, dpr[cur + i]) for i in range(depths[2])])self.norm3 = nn.LayerNorm(embed_dims[2])cur += depths[2]self.block4 = nn.ModuleList([Block(embed_dims[3], 8, 1, 4, dpr[cur + i]) for i in range(depths[3])])self.norm4 = nn.LayerNorm(embed_dims[3])def forward(self, x: Tensor) -> Tensor:B = x.shape[0]# stage 1x, H, W = self.patch_embed1(x)for blk in self.block1:x = blk(x, H, W)x1 = self.norm1(x).reshape(B, H, W, -1).permute(0, 3, 1, 2)# stage 2x, H, W = self.patch_embed2(x1)for blk in self.block2:x = blk(x, H, W)x2 = self.norm2(x).reshape(B, H, W, -1).permute(0, 3, 1, 2)# stage 3x, H, W = self.patch_embed3(x2)for blk in self.block3:x = blk(x, H, W)x3 = self.norm3(x).reshape(B, H, W, -1).permute(0, 3, 1, 2)# stage 4x, H, W = self.patch_embed4(x3)for blk in self.block4:x = blk(x, H, W)x4 = self.norm4(x).reshape(B, H, W, -1).permute(0, 3, 1, 2)return x1, x2, x3, x4class PPM(nn.ModuleList):"""金字塔池化模型 Pyramid Pooling Modulehttps://arxiv.org/abs/1612.01105CVPR 2017年 的工作使用最大池化，获取"""def __init__(self, pool_sizes, in_channels, out_channels):super(PPM, self).__init__()self.pool_sizes = pool_sizesself.in_channels = in_channelsself.out_channels = out_channelsfor pool_size in pool_sizes:self.append(nn.Sequential(nn.AdaptiveMaxPool2d(pool_size),nn.Conv2d(self.in_channels, self.out_channels, kernel_size=1),))def forward(self, x):out_puts = []for ppm in self:ppm_out = nn.functional.interpolate(ppm(x), size=(x.size(2), x.size(3)), mode='bilinear',align_corners=True)out_puts.append(ppm_out)return out_putsclass PPMHEAD(nn.Module):def __init__(self, in_channels, out_channels, pool_sizes=[1, 2, 3, 6], ):super(PPMHEAD, self).__init__()self.pool_sizes = pool_sizesself.in_channels = in_channelsself.out_channels = out_channelsself.psp_modules = PPM(self.pool_sizes, self.in_channels, self.out_channels)self.final = nn.Sequential(nn.Conv2d(self.in_channels + len(self.pool_sizes) * self.out_channels, 4 * self.out_channels,kernel_size=1),nn.BatchNorm2d(4 * self.out_channels),nn.ReLU(),)def forward(self, x):out = self.psp_modules(x)out.append(x)out = torch.cat(out, 1)out = self.final(out)return outclass FPNHEAD(nn.Module):def __init__(self, out_channels=512, num_classes=19, channels=[64, 128, 320, 512]):"""Args:out_channels: 最后一层融合的 通道数，在分类前的通道数num_classes:  最后分类数目channels: 四层backbone的通道数"""super(FPNHEAD, self).__init__()self.num_classes = num_classesself.PPMHead = PPMHEAD(in_channels=channels[-1], out_channels=channels[-1] // 4)self.Conv_fuse1 = nn.Sequential(nn.Conv2d(channels[-2], channels[-2], 1),nn.BatchNorm2d(channels[-2]),nn.ReLU())self.Conv_fuse1_ = nn.Sequential(nn.Conv2d(channels[-2] + channels[-1], channels[-2], 1),nn.BatchNorm2d(channels[-2]),nn.ReLU())self.Conv_fuse2 = nn.Sequential(nn.Conv2d(channels[-3], channels[-3], 1),nn.BatchNorm2d(channels[-3]),nn.ReLU())self.Conv_fuse2_ = nn.Sequential(nn.Conv2d(channels[-3] + channels[-2], channels[-3], 1),nn.BatchNorm2d(channels[-3]),nn.ReLU())self.Conv_fuse3 = nn.Sequential(nn.Conv2d(channels[-4], channels[-4], 1),nn.BatchNorm2d(channels[-4]),nn.ReLU())self.Conv_fuse3_ = nn.Sequential(nn.Conv2d(channels[-4] + channels[-3], channels[-4], 1),nn.BatchNorm2d(channels[-4]),nn.ReLU())self.fuse_all = nn.Sequential(nn.Conv2d(sum(channels), out_channels, 1),nn.BatchNorm2d(out_channels),nn.ReLU())self.cls_seg = nn.Sequential(nn.Conv2d(out_channels, self.num_classes, kernel_size=3, padding=1),)def forward(self, input_fpn):"""Args:input_fpn: 四个特征图Returns:"""############################### x1 = torch.randn(1, 64, 56, 56)# x2 = torch.randn(1, 128, 28, 28)# x3 = torch.randn(1, 320, 14, 14)# x4 = torch.randn(1, 512, 7, 7)#  1/32的特征图 使用PPMHead torch.Size([1, 2048, 7, 7])# x1= [1, 512, 7, 7]x1 = self.PPMHead(input_fpn[-1])# print(x1.shape)# [1, 512, 7, 7]-->[1, 512, 14, 14]x = F.interpolate(x1,size=(x1.size(2) * 2, x1.size(3) * 2),mode='bilinear',align_corners=True)# 融合1/16的图  torch.Size([1, 3072, 14, 14])。仅仅在通道上拼接# [1, 512, 14, 14] + [1,320, 14, 14] =[1, 832, 14, 14]x = torch.cat([x, self.Conv_fuse1(input_fpn[-2])], dim=1)############################### [1, 832, 14, 14] -->[1, 320, 14, 14] ,进行通道数上的减少x2 = self.Conv_fuse1_(x)#  [1, 320, 14, 14]->[1, 320, 28,28]x = F.interpolate(x2,size=(x2.size(2) * 2, x2.size(3) * 2),mode='bilinear',align_corners=True)# 融合1/8的图# [1, 320, 28,28] +[1,  128, 28,28] = [1,  448, 28,28]x = torch.cat([x, self.Conv_fuse2(input_fpn[-3])], dim=1)# print(x.shape)############################### [1,  448, 28,28] -> [1, 128, 28, 28]进行通道上缩减。x3 = self.Conv_fuse2_(x)#  对1/8---> 1/4# [1, 128, 28, 28]-> [1, 128, 56, 56]x = F.interpolate(x3,size=(x3.size(2) * 2, x3.size(3) * 2),mode='bilinear',align_corners=True)# 融合1/4的图# [1, 128, 56, 56]+[1, 64, 56, 56]=[1, 192, 56, 56]x = torch.cat([x, self.Conv_fuse3(input_fpn[-4])], dim=1)############################### [1, 192, 56, 56]-> [1, 64, 56, 56]x4 = self.Conv_fuse3_(x)x1 = F.interpolate(x1, x4.size()[-2:], mode='bilinear', align_corners=True)x2 = F.interpolate(x2, x4.size()[-2:], mode='bilinear', align_corners=True)x3 = F.interpolate(x3, x4.size()[-2:], mode='bilinear', align_corners=True)x = self.fuse_all(torch.cat([x1, x2, x3, x4], 1))# print(x.shape)x = F.interpolate(x, size=(x.size(2) * 4, x.size(3) * 4), mode='bilinear', align_corners=True)# print(x.shape)x = self.cls_seg(x)return xclass pvt2_upernet(nn.Module):def __init__(self, num_classes, channels, size="B1"):"""类别数Args:num_classes:"""super(pvt2_upernet, self).__init__()self.backbone = PVTv2(size)self.decoder = FPNHEAD(num_classes=num_classes, channels=channels)def forward(self, x):x = self.backbone(x)x = self.decoder(x)return xdef pvt2_B1_upernet(num_classes):model = pvt2_upernet(num_classes=num_classes, size="B1", channels=[64, 128, 320, 512])return modeldef pvt2_B2_upernet(num_classes):model = pvt2_upernet(num_classes=num_classes, size="B2", channels=[64, 128, 320, 512])return modeldef pvt2_B3_upernet(num_classes):model = pvt2_upernet(num_classes=num_classes, size="B3", channels=[64, 128, 320, 512])return modeldef pvt2_B4_upernet(num_classes):model = pvt2_upernet(num_classes=num_classes, size="B3", channels=[64, 128, 320, 512])return modelif __name__ == '__main__':x=torch.randn(1,3,224,224)model=pvt2_B2_upernet(num_classes=19)y=model(x)print(y.shape)

参考资料

PVTv2: Improved Baselines with Pyramid Vision Transformer——PVT2解读 - 知乎 (zhihu.com)

Transformer主干网络——PVT_V2保姆级解析_只会git clone的程序员的博客-CSDN博客

【语义分割】2021-PVT2 CVMJ相关推荐

ICCV 2021 | 简而优：用分类器变换器进行小样本语义分割
点击上方"视学算法",选择加"星标"或"置顶" 重磅干货,第一时间送达作者 | 卢治合编辑 | 王晔本文是对发表于计算机视觉领域的顶级 ...
三种基于自监督深度估计的语义分割方法（arXiv 2021）
点击上方"3D视觉工坊",选择"星标" 干货第一时间送达作者丨泡泡机器人来源丨泡泡机器人SLAM 标题: Three Ways to Improve Sem ...
ICCV 2021 | PMF: 基于视觉感知的多传感器融合点云语义分割方法
作者丨月明星稀风萧萧@知乎来源丨https://zhuanlan.zhihu.com/p/419187044 编辑丨3D视觉工坊今天,我将分享一个 ICCV 2021 中的工作,基于视觉感知的多传 ...
ICCV 2021 | Transformer再助力！用CWT进行小样本语义分割
点击下方卡片,关注"CVer"公众号 AI/CV重磅干货,第一时间送达作者 | 卢治合编辑 | 王晔本文转载自:AI科技评论本文是对发表于计算机视觉领域的顶级会议 ICC ...
CVPR 2021 | 北大MSRA提出CPS：基于交叉伪监督的半监督语义分割
点击下方卡片,关注"CVer"公众号 AI/CV重磅干货,第一时间送达作者:Charles | 源:知乎 https://zhuanlan.zhihu.com/p/37812 ...
【文献阅读】用对比学习做弱监督语义分割（Sung-Hoon Yoon等人，ArXiv，2021）
一.背景文章题目:<Exploring Pixel-level Self-supervision for Weakly Supervised Semantic Segmentation> ...
【深度学习】语义分割：论文阅读（NeurIPS 2021）MaskFormer: per-pixel classification is not all you need
目录详情知识补充语义分割实例分割基本流程主要技术路线自上而下的实例分割方法自下而上的实例分割方法掩膜 Mask 什么是mask掩码? mask掩码有什么用? mask classif ...
CVPR2021|基于双边扩充和自适应融合方法的点云语义分割网络
Semantic Segmentation for Real Point Cloud Scenes via Bilateral Augmentation and Adaptive Fusion 1.M ...
重磅发布！Google语义分割新数据集来啦！又一个分割SOTA模型
[导读]自动驾驶里视觉一直为人所诟病,特斯拉就是经常被拉出来批判的典型.谷歌最近开发了一个新模型,效果拔群,已被CVPR2021接收. 对于人来说,看一张平面照片能够想象到重建后的3D场景布局,能够根 ...
如何用PyTorch进行语义分割？一个教程教会你｜资源
木易发自凹非寺量子位报道 | 公众号 QbitAI 很久没给大家带来教程资源啦. 正值PyTorch 1.7更新,那么我们这次便给大家带来一个PyTorch简单实用的教程资源:用PyTorc ...

【语义分割】2021-PVT2 CVMJ

文章目录

【语义分割】2021-PVT2 CVMJ

1. 简介

2. 网络

2.1 总体架构

2.2 Linear Spatial Reduction Attention

2.3 Overlapping Patch Embedding(重叠补丁嵌入)

2.4 Convolutional FeedForward

3. 代码 pvt2-upernet

【语义分割】2021-PVT2 CVMJ相关推荐

最新文章

热门文章