yolov5加入CBAM,SE,CA,ECA注意力机制,纯代码(22.3.1还更新)

本文所涉及到的yolov5网络为5.0版本,后续有需求会更新6.0版本。

CBAM注意力

# class ChannelAttention(nn.Module): # def __init__(self, in_planes, ratio=16): # super(ChannelAttention, self).__init__() # self.avg_pool = nn.AdaptiveAvgPool2d(1) # self.max_pool = nn.AdaptiveMaxPool2d(1) # # self.f1 = nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False) # self.relu = nn.ReLU() # self.f2 = nn.Conv2d(in_planes // ratio, in_planes, 1, bias=False) # # 写法二,亦可使用顺序容器 # # self.sharedMLP = nn.Sequential( # # nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False), nn.ReLU(), # # nn.Conv2d(in_planes // rotio, in_planes, 1, bias=False)) # # self.sigmoid = nn.Sigmoid() # # def forward(self, x): # avg_out = self.f2(self.relu(self.f1(self.avg_pool(x)))) # max_out = self.f2(self.relu(self.f1(self.max_pool(x)))) # out = self.sigmoid(avg_out + max_out) # return out # # # class SpatialAttention(nn.Module): # def __init__(self, kernel_size=7): # super(SpatialAttention, self).__init__() # # assert kernel_size in (3, 7), 'kernel size must be 3 or 7' # padding = 3 if kernel_size == 7 else 1 # # self.conv = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False) # self.sigmoid = nn.Sigmoid() # # def forward(self, x): # avg_out = torch.mean(x, dim=1, keepdim=True) # max_out, _ = torch.max(x, dim=1, keepdim=True) # x = torch.cat([avg_out, max_out], dim=1) # x = self.conv(x) # return self.sigmoid(x) # # # class CBAMC3(nn.Module): # # CSP Bottleneck with 3 convolutions # def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion # super(CBAMC3, self).__init__() # c_ = int(c2 * e) # hidden channels # self.cv1 = Conv(c1, c_, 1, 1) # self.cv2 = Conv(c1, c_, 1, 1) # self.cv3 = Conv(2 * c_, c2, 1) # self.m = nn.Sequential(*[Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)]) # self.channel_attention = ChannelAttention(c2, 16) # self.spatial_attention = SpatialAttention(7) # # # self.m = nn.Sequential(*[CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)]) # # def forward(self, x): # out = self.channel_attention(x) * x # print('outchannels:{}'.format(out.shape)) # out = self.spatial_attention(out) * out # return out

CBAM代码 2022.1.26更新

受大佬指点,指出上述cbam模块不匹配yolov5工程代码,yolov5加入cbam注意力的代码以下述代码为准:(如果用这段代码,yolo.py和yaml文件中相应的CBAMC3也要换成CBAM,下面的SE同理)

class ChannelAttention(nn.Module): def __init__(self, in_planes, ratio=16): super(ChannelAttention, self).__init__() self.avg_pool = nn.AdaptiveAvgPool2d(1) self.max_pool = nn.AdaptiveMaxPool2d(1) self.f1 = nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False) self.relu = nn.ReLU() self.f2 = nn.Conv2d(in_planes // ratio, in_planes, 1, bias=False) # 写法二,亦可使用顺序容器 # self.sharedMLP = nn.Sequential( # nn.Conv2d(in_planes, in_planes // ratio, 1, bias=False), nn.ReLU(), # nn.Conv2d(in_planes // rotio, in_planes, 1, bias=False)) self.sigmoid = nn.Sigmoid() def forward(self, x): avg_out = self.f2(self.relu(self.f1(self.avg_pool(x)))) max_out = self.f2(self.relu(self.f1(self.max_pool(x)))) out = self.sigmoid(avg_out + max_out) return out class SpatialAttention(nn.Module): def __init__(self, kernel_size=7): super(SpatialAttention, self).__init__() assert kernel_size in (3, 7), 'kernel size must be 3 or 7' padding = 3 if kernel_size == 7 else 1 self.conv = nn.Conv2d(2, 1, kernel_size, padding=padding, bias=False) self.sigmoid = nn.Sigmoid() def forward(self, x): avg_out = torch.mean(x, dim=1, keepdim=True) max_out, _ = torch.max(x, dim=1, keepdim=True) x = torch.cat([avg_out, max_out], dim=1) x = self.conv(x) return self.sigmoid(x) class CBAM(nn.Module): # CSP Bottleneck with 3 convolutions def __init__(self, c1, c2, ratio=16, kernel_size=7): # ch_in, ch_out, number, shortcut, groups, expansion super(CBAM, self).__init__() # c_ = int(c2 * e) # hidden channels # self.cv1 = Conv(c1, c_, 1, 1) # self.cv2 = Conv(c1, c_, 1, 1) # self.cv3 = Conv(2 * c_, c2, 1) # self.m = nn.Sequential(*[Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)]) self.channel_attention = ChannelAttention(c1, ratio) self.spatial_attention = SpatialAttention(kernel_size) # self.m = nn.Sequential(*[CrossConv(c_, c_, 3, 1, g, 1.0, shortcut) for _ in range(n)]) def forward(self, x): out = self.channel_attention(x) * x # print('outchannels:{}'.format(out.shape)) out = self.spatial_attention(out) * out return out

1.这里是卷积注意力的代码,我一般喜欢加在common.py的C3模块后面,不需要做改动,傻瓜ctrl+c+v就可以了。

2.在yolo.py里做改动。在parse_model函数里将对应代码用以下代码替换,还是傻瓜ctrl+c+v。

 if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3, C3TR,CBAMC3]: c1, c2 = ch[f], args[0] if c2 != no: # if not output c2 = make_divisible(c2 * gw, 8) args = [c1, c2, *args[1:]] if m in [BottleneckCSP, C3,CBAMC3]: args.insert(2, n) # number of repeats n = 1

3.在yaml文件里改动。比如你要用s网络,我是这样改的:将骨干网络中的C3模块全部替换为CBAMC3模块(这里需要注意的是,这样改动无法加载预训练权重,也就是说,你的模型是从零开始训练的)。如果不想改动这么大,那么接着往下看。

pytorch中加入注意力机制(CBAM),以yolov5为例_YY_172的博客-CSDN博客_yolov5加注意力

这是首发将CBAM注意力添加到yolov5网络中的博主,我也是看了他的方法,侵删。

backbone: # [from, number, module, args] [[-1, 1, Focus, [64, 3]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3,CBAMC3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 9, CBAMC3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, CBAMC3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 1, SPP, [1024, [5, 9, 13]]], [-1, 3, CBAMC3, [1024, False]], # 9 ]

SE注意力

class SELayer(nn.Module): def __init__(self, c1, r=16): super(SELayer, self).__init__() self.avgpool = nn.AdaptiveAvgPool2d(1) self.l1 = nn.Linear(c1, c1 // r, bias=False) self.relu = nn.ReLU(inplace=True) self.l2 = nn.Linear(c1 // r, c1, bias=False) self.sig = nn.Sigmoid() def forward(self, x): b, c, _, _ = x.size() y = self.avgpool(x).view(b, c) y = self.l1(y) y = self.relu(y) y = self.l2(y) y = self.sig(y) y = y.view(b, c, 1, 1) return x * y.expand_as(x) 

2022.1.26SE代码更新

受同一位大佬指正,上述部分的se代码同样没有匹配yolov5工程代码,将修改后的se代码贴出,se注意力的代码以下述为准:

class SE(nn.Module): def __init__(self, c1, c2, r=16): super(SE, self).__init__() self.avgpool = nn.AdaptiveAvgPool2d(1) self.l1 = nn.Linear(c1, c1 // r, bias=False) self.relu = nn.ReLU(inplace=True) self.l2 = nn.Linear(c1 // r, c1, bias=False) self.sig = nn.Sigmoid() def forward(self, x): print(x.size()) b, c, _, _ = x.size() y = self.avgpool(x).view(b, c) y = self.l1(y) y = self.relu(y) y = self.l2(y) y = self.sig(y) y = y.view(b, c, 1, 1) return x * y.expand_as(x)

1.这里是SE注意力的代码段,同上一个注意力的加法一样,我喜欢加在C3后面。

2.在yolo.py中做改动。

def parse_model(d, ch): # model_dict, input_channels(3) logger.info('n%3s%18s%3s%10s %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments')) anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors no = na * (nc + 5) # number of outputs = anchors * (classes + 5) layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args m = eval(m) if isinstance(m, str) else m # eval strings for j, a in enumerate(args): try: args[j] = eval(a) if isinstance(a, str) else a # eval strings except: pass n = max(round(n * gd), 1) if n > 1 else n # depth gain if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3, C3TR, CoordAtt, SELayer, eca_layer, CBAM]: c1, c2 = ch[f], args[0] if c2 != no: # if not output c2 = make_divisible(c2 * gw, 8) args = [c1, c2, *args[1:]] if m in [BottleneckCSP, C3, C3TR]: args.insert(2, n) # number of repeats n = 1 elif m is nn.BatchNorm2d: args = [ch[f]] elif m is Concat: c2 = sum([ch[x] for x in f]) elif m is Detect: args.append([ch[x] for x in f]) if isinstance(args[1], int): # number of anchors args[1] = [list(range(args[1] * 2))] * len(f) elif m is Contract: c2 = ch[f] * args[0] ** 2 elif m is Expand: c2 = ch[f] // args[0] ** 2 else: c2 = ch[f]

3.在你要用的yaml文件中做改动。

backbone: # [from, number, module, args] [[-1, 1, Focus, [64, 3]], # 0-P1/2 [-1, 1, Conv, [128, 3, 2]], # 1-P2/4 [-1, 3,C3, [128]], [-1, 1, Conv, [256, 3, 2]], # 3-P3/8 [-1, 9, C3, [256]], [-1, 1, Conv, [512, 3, 2]], # 5-P4/16 [-1, 9, C3, [512]], [-1, 1, Conv, [1024, 3, 2]], # 7-P5/32 [-1, 1, SPP, [1024, [5, 9, 13]]], [-1, 3, C3, [1024, False]], # 9 [-1, 1, SELayer, [1024, 4]] ]

运行成功后是这样的

应该能看到那个注意力加在哪里了吧,这就是用上了。

这是我用的另一种添加注意力的方法,这种方法会加载预训练权重,推荐大家使用这种方法。既然推荐大家使用这种方法,那我推荐添加CBAM注意力那种方法目的是啥呢?哈哈哈哈再往下看。

天池竞赛-布匹缺陷检测baseline提升过程-给yolov5模型添加注意力机制_pprp的博客-CSDN博客_yolov5注意力机制

这是我看的将SE注意力添加到 yolov5模型里的博客,我同样也是引用了这位博主的方法,感谢分享,侵删。

ECA注意力

# class eca_layer(nn.Module): # """Constructs a ECA module. # Args: # channel: Number of channels of the input feature map # k_size: Adaptive selection of kernel size # """ # def __init__(self, channel, k_size=3): # super(eca_layer, self).__init__() # self.avg_pool = nn.AdaptiveAvgPool2d(1) # self.conv = nn.Conv1d(1, 1, kernel_size=k_size, padding=(k_size - 1) // 2, bias=False) # self.sigmoid = nn.Sigmoid() # # def forward(self, x): # # feature descriptor on the global spatial information # y = self.avg_pool(x) # # # Two different branches of ECA module # y = self.conv(y.squeeze(-1).transpose(-1, -2)).transpose(-1, -2).unsqueeze(-1) # # # Multi-scale information fusion # y = self.sigmoid(y) # x=x*y.expand_as(x) # # return x * y.expand_as(x)

1.这里是注意力代码片段,放到自己的脚本里把注释取消掉就可以了,添加的位置同上,这里就不说了。

2.改动yolo.py。看以下代码段。

 if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3, C3TR]: c1, c2 = ch[f], args[0] if c2 != no: # if not output c2 = make_divisible(c2 * gw, 8) args = [c1, c2, *args[1:]] if m in [BottleneckCSP, C3,eca_layer]: args.insert(2, n) # number of repeats n = 1 elif m is nn.BatchNorm2d: args = [ch[f]] elif m is Concat: c2 = sum([ch[x] for x in f]) elif m is Detect: args.append([ch[x] for x in f]) if isinstance(args[1], int): # number of anchors args[1] = [list(range(args[1] * 2))] * len(f) elif m is Contract: c2 = ch[f] * args[0] ** 2 elif m is Expand: c2 = ch[f] // args[0] ** 2 elif m is eca_layer: channel=args[0] channel=make_divisible(channel*gw,8)if channel != no else channel args=[channel] else: c2 = ch[f]

3.改动你要用的yaml文件。这里我要解释一下为什么交代了两种添加注意力的方法(第一种:将骨干里的C3全部替换掉;第二种:在骨干最后一层加注意力,做一个输出层)。第二种方法的模型目前还在跑,还没出结果,不过模型的结果也能猜个大概,有稳定的微小提升,detect效果不会提升太多;我在用第一种方法将ECA注意力全部替换掉骨干里的C3时,模型的p、r、map均出现了下降的情况,大概就是一个两个点,但是令人意外的是,他的检测效果很好,能够检测到未作改动前的模型很多检测不到的目标,当然也会比原模型出现更多的误检和漏检情况,手动改阈值后好了很多,因为数据集涉及到公司机密,所以这里就不放出来了,我做的是安全帽的检测,有兴趣的同学可以尝试一下这种添加注意力的方法。

看下其中一张的检测结果。

如果只是求提高模型准确率,推荐第二种方法。

接下来就是发表在今年CVPR上的注意力了。

CoorAttention

# class h_sigmoid(nn.Module): # def __init__(self, inplace=True): # super(h_sigmoid, self).__init__() # self.relu = nn.ReLU6(inplace=inplace) # # def forward(self, x): # return self.relu(x + 3) / 6 # # # class h_swish(nn.Module): # def __init__(self, inplace=True): # super(h_swish, self).__init__() # self.sigmoid = h_sigmoid(inplace=inplace) # # def forward(self, x): # return x * self.sigmoid(x) # class CoordAtt(nn.Module): # def __init__(self, inp, oup, reduction=32): # super(CoordAtt, self).__init__() # self.pool_h = nn.AdaptiveAvgPool2d((None, 1)) # self.pool_w = nn.AdaptiveAvgPool2d((1, None)) # # mip = max(8, inp // reduction) # # self.conv1 = nn.Conv2d(inp, mip, kernel_size=1, stride=1, padding=0) # self.bn1 = nn.BatchNorm2d(mip) # self.act = h_swish() # # self.conv_h = nn.Conv2d(mip, oup, kernel_size=1, stride=1, padding=0) # self.conv_w = nn.Conv2d(mip, oup, kernel_size=1, stride=1, padding=0) # # def forward(self, x): # identity = x # # n, c, h, w = x.size() # x_h = self.pool_h(x) # x_w = self.pool_w(x).permute(0, 1, 3, 2) # # y = torch.cat([x_h, x_w], dim=2) # y = self.conv1(y) # y = self.bn1(y) # y = self.act(y) # # x_h, x_w = torch.split(y, [h, w], dim=2) # x_w = x_w.permute(0, 1, 3, 2) # # a_h = self.conv_h(x_h).sigmoid() # a_w = self.conv_w(x_w).sigmoid() # # out = identity * a_w * a_h # # return out

这是代码段,加在common.py的C3模块后面

这里是改动yolo.py的部分,最后在yaml文件里的改动这里就不说了,前面提供了两种方法供大家使用,大家可以自行选择。

def parse_model(d, ch): # model_dict, input_channels(3) logger.info('n%3s%18s%3s%10s %-40s%-30s' % ('', 'from', 'n', 'params', 'module', 'arguments')) anchors, nc, gd, gw = d['anchors'], d['nc'], d['depth_multiple'], d['width_multiple'] na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors no = na * (nc + 5) # number of outputs = anchors * (classes + 5) layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out for i, (f, n, m, args) in enumerate(d['backbone'] + d['head']): # from, number, module, args m = eval(m) if isinstance(m, str) else m # eval strings for j, a in enumerate(args): try: args[j] = eval(a) if isinstance(a, str) else a # eval strings except: pass n = max(round(n * gd), 1) if n > 1 else n # depth gain if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, DWConv, MixConv2d, Focus, CrossConv, BottleneckCSP, C3, C3TR,CBAMC3,CoordAtt]:# c1, c2 = ch[f], args[0] if c2 != no: # if not output c2 = make_divisible(c2 * gw, 8) args = [c1, c2, *args[1:]] if m in [BottleneckCSP, C3, C3TR]: args.insert(2, n) # number of repeats n = 1 elif m is nn.BatchNorm2d: args = [ch[f]] elif m is Concat: c2 = sum([ch[x] for x in f]) elif m is Detect: args.append([ch[x] for x in f]) if isinstance(args[1], int): # number of anchors args[1] = [list(range(args[1] * 2))] * len(f) elif m is Contract: c2 = ch[f] * args[0] ** 2 elif m is Expand: c2 = ch[f] // args[0] ** 2 # elif m is eca_layer: # channel=args[0] # channel=make_divisible(channel*gw,8)if channel != no else channel # args=[channel] elif m is CoordAtt: inp,oup,re = args[0],args[1],args[2] oup = make_divisible(oup * gw, 8) if oup != no else oup args = [inp,oup,re] else: c2 = ch[f]

后面的ECA和CA注意力添加方法是我对着前两位博主照葫芦画瓢,在我的本地运行多次,就俩字,好用,以后的注意力也可以按照这种方法去添加。

各种注意力的添加方法以及如何work,我都懂一些,如果有需要的朋友可以联系我,赚点生活费。


2022.2.14更:本人已实现使用densenet替换focus、neck中fpn结构改为bi-fpn代码,有需要的小伙伴请私聊,赚点生活费。可用于毕业以及硕士小论文发表的trick。


我和我的朋友们组建了一间小小的工作室,团队里有不乏双一流大学人工智能领域的学生,国赛排名前五的大佬,从业多年的工程师,最低学历硕士研究生,针对学生的毕业论文或者课程设计等需求会有相应的照顾;针对工程项目,我们承诺以最短的时间满足您的需求。当然,如果您是大学生或者研究生,觉得自己的能力不错,我们也会为您提供一份兼职收入,欢迎加入我们,群聊二维码如下,一周一更,谢谢大家支持。

不胜感激,最后祝大家年薪百万。

扯完了。

原文链接:https://blog.csdn.net/zqt321/article/details/121772854

更多内容