[论文速览]：Multi-source Domain Adaptation for Semantic Segmentation
Multi-source Domain Adaptation for Semantic Segmentation
[NeurIPS 2019] [2020: MADAN: Multi-source Adversarial Domain Aggregation Network for Domain Adaptation] [github]
目录
Multi-source Domain Adaptation for Semantic Segmentation
Abstract
Problem Setup
MADAN
Overview
Dynamic Adversarial Image Generation
Adversarial Domain Aggregation
Feature-aligned Semantic Segmentation
MADAN Learning
Abstract
Simulation-to-real domain adaptation for semantic segmentation has been actively studied for various applications such as autonomous driving. Existing methods mainly focus on a single-source setting, which cannot easily handle a more practical scenario of multiple sources with different distributions. In this paper, we propose to investigate multi-source domain adaptation for semantic segmentation. Specifically, we design a novel framework, termed Multi-source Adversarial Domain Aggregation Network (MADAN), which can be trained in an end-to-end manner. First, we generate an adapted domain for each source with dynamic semantic consistency while aligning at the pixel-level cycle-consistently towards the target. Second, we propose sub-domain aggregation discriminator and cross-domain cycle discriminator to make different adapted domains more closely aggregated. Finally, feature-level alignment is performed between the aggregated domain and target domain while training the segmentation network. Extensive experiments from synthetic GTA and SYNTHIA to real Cityscapes and BDDS datasets demonstrate that the proposed MADAN model outperforms state-of-the-art approaches.
第一句，研究意义：在自动驾驶等多种应用中，针对语义分割的仿真-真实域适应问题进行了研究。
第二句，提出问题：现有的方法主要集中在单一源域设置上，这很难处理具有不同分布的、多个源的、更实际的场景。
第三句，本文工作：一句话概括本文做的内容，即 提出研究语义分割的多源域适应问题。
第四-七句，算法介绍：本文提出的模型，Multi-source Adversarial Domain Aggregation Network (MADAN)，包括三个部分：
First，为每个源生成一个具有动态语义一致性的自适应域，同时在像素级循环上对齐目标；
Second，出了子域聚集判别器和跨域循环判别器，以使不同的适应域更紧密地聚集在一起；
Finally，在训练分割网络的同时，对聚集的域和目标域进行特征级对齐。
第八句，实验结论：从合成的数据集 GTA 和 SYNTHIA 到真实的数据集 Cityscapes 和 BDDS 上的广泛实验表明，所提出的 MADAN 模型优于最先进的方法。
Problem Setup
MADAN
Overview
Figure 1: The framework of the proposed Multi-source Adversarial Domain Aggregation Network (MADAN). The colored solid arrows represent generators, while the black solid arrows indicate the segmentation network F. The dashed arrows correspond to different losses.
Dynamic Adversarial Image Generation
Adversarial Domain Aggregation
Feature-aligned Semantic Segmentation
MADAN Learning
[论文速览]：Multi-source Domain Adaptation for Semantic Segmentation相关推荐
- 论文笔记-Domain Adaptation for Semantic Segmentation with Maximum Squares Loss
论文信息 论文标题:Domain Adaptation for Semantic Segmentation with Maximum Squares Loss 论文作者:Minghao Chen, H ...
- 语义分割-Unsupervised Domain Adaptation in Semantic Segmentation：a Review语义分割中的无监督领域自适应：综述
Unsupervised Domain Adaptation in Semantic Segmentation:a Review语义分割中的无监督领域自适应:综述 0.摘要 1.介绍 1.1.语义分割 ...
- Unsupervised Domain Adaptation for Semantic Segmentation by Content Transfer
论文链接:https://arxiv.org/abs/2012.12545 此篇论文被AAAI2021接收. 文章目录 1 Motivation 1.1 Domain Gap 1.2 Class Im ...
- 【弱监督学习】Weakly Supervised Adversarial Domain Adaptation for Semantic Segmentation in Urban Scenes
0. 前言 本文是2019年发表在IEEE Transactions On Image Processing期刊上的一篇关于弱监督的城市场景语义分割的文章,文章中采用两个域分类器进行对抗训练,从而提高 ...
- 2018Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training
文章目录 一.简单介绍 一.简单介绍 1)Motivation 不采用对抗的方式,也能取得比较好的适应性能,文章想把全局的,具体类别的特征对齐和任务本身用一个简单的loss统一起来 ...
- 【域自适应】Dual Path Learning for Domain Adaptation of Semantic Segmentation
文章目录 摘要 1.介绍 2.相关工作 3.方法 3.1. Single Path Warm-up 3.2. Dual Path Image Translation 3.3. Dual Path Ad ...
- 【今日CS 视觉论文速览】 27 Dec 2018
今日CS.CV计算机视觉论文速览 Thu, 27 Dec 2018 Totally 70 papers Interesting: 荧光显微镜数据集FMD,提供了包含12000张真实荧光显微镜的照片.主 ...
- 【今日CV 计算机视觉论文速览 第94期】 4 Apr 2019
今日CS.CV 计算机视觉论文速览 Thu, 4 Apr 2019 Totally 59 papers Interesting: ?点云过分割直接到superpoints,研究人员将这一问题转为了如何 ...
- 【今日CV 计算机视觉论文速览 第103期】Mon, 22 Apr 2019
今日CS.CV 计算机视觉论文速览 Mon, 22 Apr 2019 Totally 38 papers ?上期速览✈更多精彩请移步主页 Interesting: ?Fashion++, 如何通过微小 ...
最新文章
- c语言如何持续输入字符直到指定字符结束_《小白学C》第三章 常用输入输出函数...
- Delphi FastReport动态加载图片
- Windows系统笔记本禁用自带键盘
- 【风控场景】互利网上数字金融典型场景： 网络营销
- PurdueUCLA提出梯度Boosting网络，效果远好于XGBoost模型！
- mac 系统使用macaca inspector 获取iphone真机应用元素
- MATLAB（四）在高等数学中的应用
- 企业实战_18_MyCat_ZK集群安装部署
- 监听浏览器是否被缩放 - 案例篇
- 奔跑吧火柴人c语言开发_小学数学智力题：这两个火柴棍趣味题，你能做出来吗？...
- arduino蜂鸣器音乐代码青花瓷_教你用 Arduino 制造简易金属探测器
- 联想高校AI精英挑战赛总冠军出炉！助力中国迎来智能变革
- paip.提升用户体验----记住用户名与自动登录
- mysql数据库恢复
- JS图片压缩+图片上传前检测类型、大小、尺寸
- 常看：前端面试题搂搂
- [PyG] 1.如何使用GCN完成一个最基本的训练过程(含GCN实现)
- 关于导入.a文件后报错Undefined symbols for architecture arm64:
- 计算机网络总结(1)
- 5.23 C语言练习（平均值计算：输入10个整数，求它们的平均值，并输出大于平均值的数据的个数。）