1.CoTAttention

论文地址：2107.12292.pdf (arxiv.org)

CoTAttention网络是一种用于多模态场景下的视觉问答（Visual Question Answering，VQA）任务的神经网络模型。它是在经典的注意力机制（Attention Mechanism）上进行了改进，能够自适应地对不同的视觉和语言输入进行注意力分配，从而更好地完成VQA任务。

CoTAttention网络中的“CoT”代表“Cross-modal Transformer”，即跨模态Transformer。在该网络中，视觉和语言输入分别被编码为一组特征向量，然后通过一个跨模态的Transformer模块进行交互和整合。在这个跨模态的Transformer模块中，Co-Attention机制被用来计算视觉和语言特征之间的交互注意力，从而实现更好的信息交换和整合。在计算机视觉和自然语言处理紧密结合的VQA任务中，CoTAttention网络取得了很好的效果。

新加坡国立大学的Qibin Hou等人提出了一种为轻量级网络设计的新的注意力机制，该机制将位置信息嵌入到了通道注意力中，称为coordinate attention

京东AI Research提出新的主干网络CoTNet,在CVPR上2021获得开放域图像识别竞赛冠军

京东AI Research提出一种Contextual Transformer Networks结构，就是将Transformer捕捉全局信息的能力与CNN捕捉临近局部信息能力相结合，从而提高网络模型的特征表达能力。值得注意的是，该方法可以实现模块的“即插即用”，将ResNet网络中的3x3模块替换成CoTNet的核心模块即可使用，Res2Net网络也是基于这种思想实现的。

在相同深度（50层或101层）下，top-1和top-5结果都表明本文的方法比卷积网络和Attention-based网络性能更好。

2. 基于Yolov8的CoordAttention实现

2.1 CoTAttention 加入 modules.py中

核心代码：

######################  CoTAttention   ####     start   by  AI&CV  ###############################
import torch
from torch import flatten, nn
from torch.nn import functional as Fclass CoTAttention(nn.Module):def __init__(self, dim=512, kernel_size=3):super().__init__()self.dim = dimself.kernel_size = kernel_sizeself.key_embed = nn.Sequential(nn.Conv2d(dim, dim, kernel_size=kernel_size, padding=kernel_size // 2, groups=4, bias=False),nn.BatchNorm2d(dim),nn.ReLU())self.value_embed = nn.Sequential(nn.Conv2d(dim, dim, 1, bias=False),nn.BatchNorm2d(dim))factor = 4self.attention_embed = nn.Sequential(nn.Conv2d(2 * dim, 2 * dim // factor, 1, bias=False),nn.BatchNorm2d(2 * dim // factor),nn.ReLU(),nn.Conv2d(2 * dim // factor, kernel_size * kernel_size * dim, 1))def forward(self, x):bs, c, h, w = x.shapek1 = self.key_embed(x)  # bs,c,h,wv = self.value_embed(x).view(bs, c, -1)  # bs,c,h,wy = torch.cat([k1, x], dim=1)  # bs,2c,h,watt = self.attention_embed(y)  # bs,c*k*k,h,watt = att.reshape(bs, c, self.kernel_size * self.kernel_size, h, w)att = att.mean(2, keepdim=False).view(bs, c, -1)  # bs,c,h*wk2 = F.softmax(att, dim=-1) * vk2 = k2.view(bs, c, h, w)return k1 + k2######################  CoTAttention   ####     end   by  AI&CV  ###############################

​2.2 yolov8_CoTAttention.yaml

# Ultralytics YOLO  , GPL-3.0 license
# YOLOv8 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect# Parameters
nc: 4  # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov8n.yaml' will call yolov8.yaml with scale 'n'# [depth, width, max_channels]n: [0.33, 0.25, 1024]  # YOLOv8n summary: 225 layers,  3157200 parameters,  3157184 gradients,   8.9 GFLOPss: [0.33, 0.50, 1024]  # YOLOv8s summary: 225 layers, 11166560 parameters, 11166544 gradients,  28.8 GFLOPsm: [0.67, 0.75, 768]   # YOLOv8m summary: 295 layers, 25902640 parameters, 25902624 gradients,  79.3 GFLOPsl: [1.00, 1.00, 512]   # YOLOv8l summary: 365 layers, 43691520 parameters, 43691504 gradients, 165.7 GFLOPsx: [1.00, 1.25, 512]   # YOLOv8x summary: 365 layers, 68229648 parameters, 68229632 gradients, 258.5 GFLOPs# YOLOv8.0n backbone
backbone:# [from, repeats, module, args]- [-1, 1, Conv, [64, 3, 2]]  # 0-P1/2- [-1, 1, Conv, [128, 3, 2]]  # 1-P2/4- [-1, 3, C2f, [128, True]]- [-1, 1, Conv, [256, 3, 2]]  # 3-P3/8- [-1, 6, C2f, [256, True]]- [-1, 1, Conv, [512, 3, 2]]  # 5-P4/16- [-1, 6, C2f, [512, True]]- [-1, 1, Conv, [1024, 3, 2]]  # 7-P5/32- [-1, 3, C2f, [1024, True]]- [-1, 1, SPPF, [1024, 5]]  # 9# YOLOv8.0n head
head:- [-1, 1, nn.Upsample, [None, 2, 'nearest']]- [[-1, 6], 1, Concat, [1]]  # cat backbone P4- [-1, 3, C2f, [512]]  # 12- [-1, 1, nn.Upsample, [None, 2, 'nearest']]- [[-1, 4], 1, Concat, [1]]  # cat backbone P3- [-1, 3, C2f, [256]]  # 15 (P3/8-small)- [-1, 3, CoTAttention, [256]]   # 16- [-1, 1, Conv, [256, 3, 2]]- [[-1, 12], 1, Concat, [1]]  # cat head P4- [-1, 3, C2f, [512]]  # 19 (P4/16-medium)- [-1, 3, CoTAttention, [512]]  - [-1, 1, Conv, [512, 3, 2]]- [[-1, 9], 1, Concat, [1]]  # cat head P5- [-1, 3, C2f, [1024]]  # 23 (P5/32-large)- [-1, 3, CoTAttention, [1024]]  - [[16, 20, 24], 1, Detect, [nc]]  # Detect(P3, P4, P5)