Frequency-Based Optimal Style Mix for Domain Generalization in Semantic Segmentation of Remote Sensing Images

计算机科学 分割 人工智能 一般化 一致性(知识库) 频域 正规化(语言学) 领域(数学分析) 试验数据 模式识别(心理学) 算法 计算机视觉 数学 数学分析 程序设计语言
作者
Reo Iizuka,Junshi Xia,Naoto Yokoya
出处
期刊:IEEE Transactions on Geoscience and Remote Sensing [Institute of Electrical and Electronics Engineers]
卷期号:62: 1-14 被引量:50
标识
DOI:10.1109/tgrs.2023.3344670
摘要

Supervised learning methods assume that training and test data are sampled from the same distribution. However, this assumption is not always satisfied in practical situations of land cover semantic segmentation when models trained in a particular source domain are applied to other regions. This is because domain shifts caused by variations in location, time, and sensor alter the distribution of images in the target domain from that of the source domain, resulting in significant degradation of model performance. To mitigate this limitation, domain generalization (DG) has gained attention as a way of generalizing from source domain features to unseen target domains. One approach is style randomization (SR), which enables models to learn domain-invariant features through randomizing styles of images in the source domain. Despite its potential, existing methods face several challenges, such as inflexible frequency decomposition, high computational and data preparation demands, slow speed of randomization, and lack of consistency in learning. To address these limitations, we propose a frequency-based optimal style mix (FOSMix), which consists of three components: 1) full mix (FM) enhances the data space by maximally mixing the style of reference images into the source domain; 2) optimal mix (OM) keeps the essential frequencies for segmentation and randomizes others to promote generalization; and 3) regularization of consistency ensures that the model can stably learn different images with the same semantics. Extensive experiments that require the model’s generalization ability, with domain shift caused by variations in regions and resolutions, demonstrate that the proposed method achieves superior segmentation in remote sensing. The source code is available athttps://github.com/Reo-I/FOSMix.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
宝宝完成签到,获得积分10
刚刚
负责雨安完成签到 ,获得积分10
刚刚
桐桐的应助被yan采纳,获得10
刚刚
微笑梦易完成签到,获得积分10
刚刚
刚刚
刚刚
1秒前
阿萨德完成签到,获得积分10
1秒前
2秒前
2秒前
风趣的雪柳完成签到,获得积分10
2秒前
2秒前
哦吼吼吼吼完成签到 ,获得积分10
2秒前
gm发布了新的文献求助10
3秒前
小小完成签到,获得积分10
3秒前
爆米花的应助被raoarao采纳,获得40
3秒前
情怀的应助被给里奥采纳,获得10
3秒前
哈哈哈哈完成签到,获得积分10
3秒前
标致青雪发布了新的文献求助10
3秒前
小余发布了新的文献求助10
3秒前
彭于晏的应助被针地很不戳采纳,获得30
4秒前
4秒前
跳跃煎蛋发布了新的文献求助10
4秒前
冷傲的不平完成签到,获得积分20
4秒前
能干灯泡完成签到,获得积分10
4秒前
刘大宝发布了新的文献求助10
5秒前
秋风的应助被Liah采纳,获得20
5秒前
ram999完成签到 ,获得积分10
6秒前
小柔美弱完成签到,获得积分10
6秒前
SciGPT的应助被hzz采纳,获得10
7秒前
维C续航中发布了新的文献求助10
7秒前
7秒前
7秒前
7秒前
迪迪张发布了新的文献求助10
7秒前
7秒前
健壮的听兰完成签到,获得积分10
7秒前
虎皮猫大人完成签到,获得积分10
8秒前
8秒前
杏仁发布了新的文献求助10
8秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Social Psychology in the Real World 600
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8181014
求助须知:如何正确求助?哪些是违规求助? 9571039
关于积分的说明 21816042
捐赠科研通 7777369
什么是DOI,文献DOI怎么找? 3433378
关于科研通互助平台的介绍 2557621
邀请新用户注册赠送积分活动 2479385