Learning Spatial Regularization with Image-level Supervisions for Multi-label Image Classification

计算机科学 人工智能 空间关系 利用 模式识别(心理学) 正规化(语言学) 上下文图像分类 图像(数学) 人工神经网络 一般化 深层神经网络 机器学习 数学 计算机安全 数学分析
作者
Feng Zhu,Hongsheng Li,Wanli Ouyang,Nenghai Yu,Xiaogang Wang
出处
期刊:Cornell University - arXiv [Cornell University]
被引量:7
标识
DOI:10.48550/arxiv.1702.05891
摘要

Multi-label image classification is a fundamental but challenging task in computer vision. Great progress has been achieved by exploiting semantic relations between labels in recent years. However, conventional approaches are unable to model the underlying spatial relations between labels in multi-label images, because spatial annotations of the labels are generally not provided. In this paper, we propose a unified deep neural network that exploits both semantic and spatial relations between labels with only image-level supervisions. Given a multi-label image, our proposed Spatial Regularization Network (SRN) generates attention maps for all labels and captures the underlying relations between them via learnable convolutions. By aggregating the regularized classification results with original results by a ResNet-101 network, the classification performance can be consistently improved. The whole deep neural network is trained end-to-end with only image-level annotations, thus requires no additional efforts on image annotations. Extensive evaluations on 3 public datasets with different types of labels show that our approach significantly outperforms state-of-the-arts and has strong generalization capability. Analysis of the learned SRN model demonstrates that it can effectively capture both semantic and spatial relations of labels for improving classification performance.

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
JamesPei应助伶俐的傲白采纳,获得10
刚刚
Techmarine完成签到,获得积分10
刚刚
yjq完成签到,获得积分10
刚刚
如初完成签到,获得积分20
刚刚
曦_完成签到,获得积分10
刚刚
1秒前
谦让代芙完成签到,获得积分10
1秒前
科研通AI6.4应助peng采纳,获得30
1秒前
蹈死不顾完成签到,获得积分10
1秒前
组内边缘者完成签到 ,获得积分10
1秒前
2秒前
汤汤杨杨完成签到,获得积分10
2秒前
潇潇暮雨完成签到,获得积分10
2秒前
科研通AI6.4应助wyz采纳,获得10
3秒前
童行发布了新的文献求助10
4秒前
沉默天宇发布了新的文献求助10
4秒前
CJ发布了新的文献求助10
5秒前
灵魂歌手完成签到,获得积分10
5秒前
XWLi完成签到,获得积分10
5秒前
lemontea完成签到,获得积分10
6秒前
直率的大楚完成签到,获得积分10
6秒前
叮叮叮发布了新的文献求助10
6秒前
8秒前
8秒前
言庭兰玉完成签到,获得积分10
8秒前
9秒前
君莫笑完成签到,获得积分10
9秒前
让地球种满香菜完成签到,获得积分10
9秒前
9秒前
edmund完成签到,获得积分10
10秒前
10秒前
橘子海完成签到 ,获得积分10
10秒前
英姑应助ddddd采纳,获得10
11秒前
11秒前
桐桐应助ZZzz采纳,获得10
11秒前
11秒前
11秒前
12秒前
wanci应助zhogwe采纳,获得10
12秒前
12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
Autoparametric Resonance in Mechanical Systems 1000
Effects of Two Weeks of Red Light Therapy on Choroidal Thickness and Axial Length in Young Adults 700
Cosmos as Art Object: Studies in Plato's Timaeus and Other Dialogues 600
Management and the Arts 510
Matrix Methods in Data Mining and Pattern Recognition Second Edition 510
Common Foundations of American and East Asian Modernisation: From Alexander Hamilton to Junichero Koizumi 400
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 工程类 有机化学 化学工程 生物化学 计算机科学 内科学 物理 复合材料 催化作用 细胞生物学 无机化学 光电子学 物理化学 电极 基因
热门帖子
关注 科研通微信公众号,转发送积分 7670182
求助须知:如何正确求助?哪些是违规求助? 9237967
关于积分的说明 19890999
捐赠科研通 7239574
什么是DOI,文献DOI怎么找? 3284628
关于科研通互助平台的介绍 2443157
邀请新用户注册赠送积分活动 2286565