Meta-DETR: Image-Level Few-Shot Detection with Inter-Class Correlation Exploitation

计算机科学 人工智能 一般化 班级(哲学) 相关性 杠杆(统计) 机器学习 目标检测 数据挖掘 模式识别(心理学) 数学 数学分析 几何学
作者
Gongjie Zhang,Zhipeng Luo,Kaiwen Cui,Shijian Lu,Eric P. Xing
出处
期刊:IEEE Transactions on Pattern Analysis and Machine Intelligence [Institute of Electrical and Electronics Engineers]
卷期号:: 1-12 被引量:76
标识
DOI:10.1109/tpami.2022.3195735
摘要

Few-shot object detection has been extensively investigated by incorporating meta-learning into region-based detection frameworks. Despite its success, the said paradigm is still constrained by several factors, such as (i) low-quality region proposals for novel classes and (ii) negligence of the inter-class correlation among different classes. Such limitations hinder the generalization of base-class knowledge for the detection of novel-class objects. In this work, we design Meta-DETR, which (i) is the first image-level few-shot detector, and (ii) introduces a novel inter-class correlational meta-learning strategy to capture and leverage the correlation among different classes for robust and accurate few-shot object detection. Meta-DETR works entirely at image level without any region proposals, which circumvents the constraint of inaccurate proposals in prevalent few-shot detection frameworks. In addition, the introduced correlational meta-learning enables Meta-DETR to simultaneously attend to multiple support classes within a single feedforward, which allows to capture the inter-class correlation among different classes, thus significantly reducing the misclassification over similar classes and enhancing knowledge generalization to novel classes. Experiments over multiple few-shot object detection benchmarks show that the proposed Meta-DETR outperforms state-of-the-art methods by large margins. The implementation codes are available at https://github.com/ZhangGongjie/Meta-DETR.
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
刚刚
欣欣完成签到 ,获得积分10
2秒前
休比里斯老板完成签到,获得积分10
3秒前
毛豆应助哭泣的采波采纳,获得10
9秒前
9秒前
chen0424完成签到,获得积分10
11秒前
英俊的铭应助xxx采纳,获得10
11秒前
13秒前
坚强莺发布了新的文献求助10
13秒前
zsy完成签到,获得积分10
13秒前
PHW完成签到,获得积分10
14秒前
跳跃仙人掌应助仙人掌采纳,获得20
14秒前
酷波er应助沐曦采纳,获得10
15秒前
NaCl完成签到 ,获得积分10
15秒前
leezhen完成签到,获得积分10
15秒前
斯文败类应助liyong采纳,获得10
15秒前
...发布了新的文献求助50
16秒前
18秒前
19秒前
19秒前
19秒前
乐乐应助七七七呀采纳,获得10
21秒前
标致的泥猴桃完成签到,获得积分10
22秒前
22秒前
xxx发布了新的文献求助10
24秒前
24秒前
砂锅粥发布了新的文献求助10
25秒前
顺时针完成签到,获得积分10
26秒前
design完成签到,获得积分10
26秒前
乐乐应助LabRat采纳,获得10
27秒前
liyong发布了新的文献求助10
28秒前
JamesPei应助温暖幻桃采纳,获得10
29秒前
31秒前
背后书双完成签到 ,获得积分10
32秒前
小二郎应助南华知识分子采纳,获得10
32秒前
wanci应助优秀老师采纳,获得10
33秒前
33秒前
小菜鸟001完成签到,获得积分10
36秒前
37秒前
38秒前
高分求助中
Licensing Deals in Pharmaceuticals 2019-2024 3000
Cognitive Paradigms in Knowledge Organisation 2000
Effect of reactor temperature on FCC yield 2000
Introduction to Spectroscopic Ellipsometry of Thin Film Materials Instrumentation, Data Analysis, and Applications 1200
How Maoism Was Made: Reconstructing China, 1949-1965 800
Medical technology industry in China 600
ANSYS Workbench基础教程与实例详解 510
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 3312139
求助须知:如何正确求助?哪些是违规求助? 2944769
关于积分的说明 8521299
捐赠科研通 2620463
什么是DOI,文献DOI怎么找? 1432849
科研通“疑难数据库(出版商)”最低求助积分说明 664797
邀请新用户注册赠送积分活动 650115