Dual attention network for scene segmentation

设计巧妙，但是没法解释方法的有效性

dual attention的思路太多了，这篇是应用在场景语义分割上。

是一种channel-wise attention和spital-attention整合结构设计

channel-wise attention(卷积通道注意力): senet，卷积网络每一层有很多卷积核，每一个卷积核对应一个特征通道，相对于空间注意力机制，通道注意力在于分配各个卷积通道之间的资源。

spatial attention(视觉空间注意力机制):就是对于特征图上的每个位置进行attention调整，使模型关注到值得关注的区域上。

Abstract: self-attention mechanism. A dual attention networks to adaptively integeate local features with their global dependencies.运用了两种attention结构，在空间和通道维度上。

The position attention module selectively aggregates the features at each position by a weighted sum of the features at all positions. Similar features would be related to each other regardless of their distances.

The channel attention module selectively emphasizes interdependent channel maps by integrating associated features among all channel maps

1. Introduction

提出了DANet去抓住全局特征依赖在空间和通道维度，

PAM和CAM

完成了很高的精度

2. Related work

Semantic segmentation

Attention modules

3. Dual attention network

Position Attention Module

语义关系对场景理解来讲很重要，目的在于抓住全局依赖性而无关位置。然而许多工作表明，局部特征信息不够。In order to model rich contectual dependencies over local feature representations, we introducte a PAM.

The PAM encodes a wider range of contextual information into local features, thus enhancing their representive capability. Next, we elaborate the process to adaptively aggregate spatia contexts.

B,C,D经过conv，bn，relu得到三组feature map，分别reshape成C*N（c*h*w,h*w=N）,B转置N*C,B,C相乘，N*C*C*N得到N*N（H*W*H*W），经过softmax出来每个position都是0-1之间，N*N成为S，soatial attention map，

（Note that the more similar feature representations of the two position contributes to greater correlation between them）D和S相乘（C*N*N*N）为C*N，在reshape成C*H*W，和orign map相加，得到E

中间这个邻接矩阵确实是人为的，不像SE中的两个fc那样。

Channel Attention Module

Each channel map of high level features can be regarded as a class-specific response,and differenr semantic responses are associated with each other.

Channel attention和前面一样，只是中间的N*N变成了C*C

Attention Module Embedding with networks

4. Experiments

Dual attention network for scene segmentation相关推荐

【语义分割】DANet Dual Attention Network for Scene Segmentation
DANet(Dual Attention Network for Scene Segmentation)在语义分割领域多个数据集上取得了STOA的结果,值得大家关注. [废话两段] 由于之前没跑过语义 ...
文章阅读：Dual Attention Network for Scene Segmentation
原文地址:Dual Attention Network for Scene Segmentation 概述多尺度上下文融合的问题: 虽然上下文融合有助于捕获不同比例的对象,但它无法利用全局视图中对象 ...
Pyramid Attention Network for Semantic Segmentation
翻译 | 林椿眄出品 | 人工智能头条(公众号ID:AI_Thinker) 近日,北京理工大学.旷视科技.北京大学联手,发表了一篇名为 Pyramid Attention Network for S ...
Hybrid Multiple Attention Network for Semantic Segmentation in Aerial Images
论文阅读: Hybrid Multiple Attention Network for Semantic Segmentation in Aerial Images 作者声明版权声明:本文为博主原创 ...
论文Pyramid Attention Network for Semantic Segmentation笔记
该论文发表于2017CVPR,由北理工+旷视科技+北大共同完成论文传送门本人主要借鉴其中的思想在目标检测中的可行性,故不考虑语义分割相关的内容初次实现,如有不足之处,还请指出,谢谢! 名词解释 ...
[论文阅读] Shallow Attention Network for Polyp Segmentation
论文地址:https://arxiv.org/abs/2108.00882 代码:https://github.com/weijun88/SANet 发表于:MICCAI'21 Abstract 准确 ...
双重关系感知注意力机制 Dual Relation-Aware Attention[keras实现 dual attention优化版]
文章目录前言一.Compat Position Attention Module紧凑型位置注意力模块二.Compat Channel Attention Module紧凑型通道注意力模块三.效 ...
【Attention】Dual Attention（DANet） Fully Attention（FLA）
空间注意力有助于保留细节信息,通道注意力有助于保留大物体的语义一致性. 有效使用两种注意力可以提升性能. 本文旨在记录一些常用的注意力,以及代码实现,包括两篇文章,DANet,FLA. Dual At ...
【论文笔记】DR-TANet: Dynamic Receptive Temporal Attention Network for Street Scene Change Detection
论文论文题目:DR-TANet: Dynamic Receptive Temporal Attention Network for Street Scene Change Detection 收录: ...

Dual attention network for scene segmentation

Dual attention network for scene segmentation相关推荐

最新文章

热门文章