ChangeMask: Deep multi-task encoder-transformer-decoder architecture for semantic change detection

计算机科学 编码器 变更检测 人工智能 变压器 判别式 自然语言处理 模式识别(心理学) 量子力学 操作系统 物理 电压
作者
Zhuo Zheng,Yanfei Zhong,Shiqi Tian,Ailong Ma,Liangpei Zhang
出处
期刊:Isprs Journal of Photogrammetry and Remote Sensing 卷期号:183: 228-239 被引量:131
标识
DOI:10.1016/j.isprsjprs.2021.10.015
摘要

Multi-temporal high spatial resolution earth observation makes it possible to detect complex urban land surface changes, which is a significant and challenging task in remote sensing communities. Previous works mainly focus on binary change detection (BCD) based on modern technologies, e.g., deep fully convolutional network (FCN), whereas the deep network architecture for semantic change detection (SCD) is insufficiently explored in current literature. In this paper, we propose a deep multi-task encoder-transformer-decoder architecture (ChangeMask) designed by exploring two important inductive biases: sematic-change causal relationship and temporal symmetry. ChangeMask decouples the SCD into a temporal-wise semantic segmentation and a BCD, and then integrates these two tasks into a general encoder-transformer-decoder framework. In the encoder part, we design a semantic-aware encoder to model the semantic-change causal relationship. This encoder is only used to learn semantic representation and then learn change representation from semantic representation via a later transformer module. In this way, change representation can constrain semantic representation during training, which introduces a regularization to reduce the risk of overfitting. To learn a robust change representation from semantic representation, we propose a temporal-symmetric transformer (TST) to guarantee temporal symmetry for change representation and keep it discriminative. Based on the above semantic representation and change representation, we adopt simple multi-task decoders to output semantic change map. Benefiting from the differentiable building blocks, ChangeMask can be trained by a multi-task loss function, which significantly simplifies the whole pipeline of applying ChangeMask. The comprehensive experimental results on two large-scale SCD datasets confirm the effectiveness and superiority of ChangeMask in SCD. Besides, to demonstrate the potential value in real-world applications, e.g., automatic urban analysis and decision-making, we deploy the ChangeMask to map a large geographic area covering 30 km2 with 300 million pixels. Code will be made available.
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
白金之星完成签到 ,获得积分10
9秒前
10秒前
12秒前
xxx7749发布了新的文献求助10
14秒前
在水一方应助称心寒松采纳,获得10
14秒前
kk发布了新的文献求助10
15秒前
15秒前
17秒前
ssz发布了新的文献求助10
21秒前
kk完成签到,获得积分10
21秒前
暮沐晓光完成签到,获得积分10
22秒前
北风应助虾米采纳,获得10
23秒前
朝闻道完成签到 ,获得积分10
25秒前
潘贤铖关注了科研通微信公众号
26秒前
27秒前
愉快天亦完成签到,获得积分10
31秒前
31秒前
羊白筠发布了新的文献求助10
32秒前
qiao应助xxx7749采纳,获得10
34秒前
yuaner发布了新的文献求助10
35秒前
36秒前
pluto应助李向东采纳,获得50
36秒前
称心寒松发布了新的文献求助10
41秒前
哭泣的赛凤完成签到 ,获得积分10
43秒前
毕业完成签到,获得积分10
47秒前
51秒前
lqqq完成签到 ,获得积分10
53秒前
学术智子完成签到,获得积分10
54秒前
56秒前
sky完成签到 ,获得积分10
56秒前
腼腆的恶天完成签到,获得积分10
56秒前
隐形傲霜完成签到 ,获得积分10
56秒前
58秒前
59秒前
1分钟前
CodeCraft应助jinzhen采纳,获得10
1分钟前
北风应助称心寒松采纳,获得10
1分钟前
zzzzz发布了新的文献求助10
1分钟前
玛卡巴卡完成签到 ,获得积分10
1分钟前
1分钟前
高分求助中
【此为提示信息,请勿应助】请按要求发布求助,避免被关 20000
Technologies supporting mass customization of apparel: A pilot project 450
Mixing the elements of mass customisation 360
Периодизация спортивной тренировки. Общая теория и её практическое применение 310
the MD Anderson Surgical Oncology Manual, Seventh Edition 300
Nucleophilic substitution in azasydnone-modified dinitroanisoles 300
Political Ideologies Their Origins and Impact 13th Edition 260
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 物理 生物化学 纳米技术 计算机科学 化学工程 内科学 复合材料 物理化学 电极 遗传学 量子力学 基因 冶金 催化作用
热门帖子
关注 科研通微信公众号,转发送积分 3781313
求助须知:如何正确求助?哪些是违规求助? 3326832
关于积分的说明 10228480
捐赠科研通 3041848
什么是DOI,文献DOI怎么找? 1669603
邀请新用户注册赠送积分活动 799153
科研通“疑难数据库(出版商)”最低求助积分说明 758751