PCLDet: Prototypical Contrastive Learning for Fine-Grained Object Detection in Remote Sensing Images

计算机科学 人工智能 目标检测 判别式 特征(语言学) 对象(语法) 特征提取 计算机视觉 特征学习 班级(哲学) 深度学习 模式识别(心理学) 遥感 语言学 地质学 哲学
作者
Lihan Ouyang,Guangmiao Guo,Leyuan Fang,Pedram Ghamisi,Jun Yue
出处
期刊:IEEE Transactions on Geoscience and Remote Sensing [Institute of Electrical and Electronics Engineers]
卷期号:61: 1-11 被引量:7
标识
DOI:10.1109/tgrs.2023.3290091
摘要

The capacity of satellites to supply high-resolution imaging has promoted the fine-grained object detection task in remote sensing images. However, this type of object detection is challenging due to low interclass feature differences in objects. To address this issue, we propose a prototypical contrastive learning-based detector (PCLDet) for fine-grained object detection in remote sensing images. The PCLDet first introduces the prototype to learn the fine-grained objects’ features, and then adopts contrastive learning to compare the target and the learned features, thus improving the differentiability of the fine-grained object. Specifically, we first introduce the prototype, which represents the feature centers of each class, and then construct a prototype bank to store the feature prototypes of each class. Then, we introduce contrastive learning to extract the discriminative features by maximizing the interclass distance and minimizing the intraclass distance. Furthermore, we propose the ProtoCL loss as a part of the model optimization, which enables more representative prototypes to be learned. Finally, to address the long-tail problem in the remote sensing fine-grained object detection dataset, we propose a new proposal sampler, the class-balanced sampler (CBS) that can sample each class equally. Extensive experiments demonstrate that our method can achieve state-of-the-art performance on a commonly used aerial fine-grained object dataset (Fair1M) and aerial fine-grained ship dataset (OFSD) while maintaining high efficiency. The code will be available at https://github.com/G-Naughty/PCLDet.
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
更新
大幅提高文件上传限制,最高150M (2024-4-1)

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
orixero应助科研通管家采纳,获得10
刚刚
香蕉觅云应助科研通管家采纳,获得10
刚刚
英姑应助科研通管家采纳,获得10
刚刚
上官若男应助科研通管家采纳,获得10
刚刚
刚刚
汉堡包应助科研通管家采纳,获得10
刚刚
在水一方应助科研通管家采纳,获得10
刚刚
打打应助科研通管家采纳,获得30
1秒前
在水一方应助科研通管家采纳,获得10
1秒前
大个应助科研通管家采纳,获得10
1秒前
情怀应助科研通管家采纳,获得10
1秒前
兀兀应助科研通管家采纳,获得10
1秒前
1秒前
Lucas应助科研通管家采纳,获得10
1秒前
1秒前
Ava应助科研通管家采纳,获得10
1秒前
1秒前
上官若男应助科研通管家采纳,获得10
1秒前
Akim应助天边一阵风采纳,获得10
2秒前
2秒前
单薄飞荷完成签到,获得积分10
2秒前
zzzyyyuuu完成签到 ,获得积分10
3秒前
4秒前
轩贝发布了新的文献求助20
5秒前
Ye完成签到 ,获得积分10
6秒前
带头大哥应助萧水白采纳,获得100
7秒前
7秒前
blue2021发布了新的文献求助10
7秒前
lily发布了新的文献求助30
7秒前
8秒前
YY完成签到,获得积分10
8秒前
9秒前
9秒前
薄荷糖发布了新的文献求助10
12秒前
自然沛槐发布了新的文献求助30
13秒前
科研通AI2S应助blue2021采纳,获得10
13秒前
13秒前
panpan发布了新的文献求助10
14秒前
单薄天蓉完成签到,获得积分10
14秒前
徐长卿完成签到 ,获得积分10
15秒前
高分求助中
Effect of reactor temperature on FCC yield 2000
Production Logging: Theoretical and Interpretive Elements 1500
Very-high-order BVD Schemes Using β-variable THINC Method 1000
PraxisRatgeber: Mantiden: Faszinierende Lauerjäger 800
Uncertainty Quantification: Theory, Implementation, and Applications, Second Edition 800
錢鍾書楊絳親友書札 600
A new species of Coccus (Homoptera: Coccoidea) from Malawi 500
热门求助领域 (近24小时)
化学 医学 生物 材料科学 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 催化作用 物理化学 免疫学 量子力学 细胞生物学
热门帖子
关注 科研通微信公众号,转发送积分 3288834
求助须知:如何正确求助?哪些是违规求助? 2926086
关于积分的说明 8425326
捐赠科研通 2597126
什么是DOI,文献DOI怎么找? 1417020
科研通“疑难数据库(出版商)”最低求助积分说明 659556
邀请新用户注册赠送积分活动 642000