当前位置:网站首页>【Random Erasing】《Random Erasing Data Augmentation》
【Random Erasing】《Random Erasing Data Augmentation》
2022-07-02 06:26:00 【bryant_meng】

AAAI-2020
文章目录
1 Background and Motivation
为提升 CNN 模型的泛化能力,提升 CNN 模型应对遮挡场景的鉴别能力,提出了一种数据增广方法,Random Erasing——random position with random-sized mask with random pixel
While we can manually add occluded natural images to the training set, it is costly and the levels of occlusion might be limited.
2 Related Work
Dropout / DropConect / Adaptive dropout / Stochastic Pooling / DisturbLabel / PatchShuffle
Random flipping / random cropping
3 Advantages / Contributions
提出 Random Erasing Data Augmentation,与 random cropping,random filpping 互补,not require any extra parameter learning,在分类 / 检测 / reID 任务上均有不错的效果
4 Method
1)算法流程
x e x_e xe 和 y e y_e ye 是中心点, W e W_e We 和 H e H_e He 是长宽
随机中心点,随机长宽比,随机面积,随机像素填充值
2)分类和 ReID 上的应用

简单粗暴,全图范围 random
3)目标检测上的应用
全图范围,目标范围,全图 + 目标范围
4)和 Random cropping 的区别
random cropping,可以降低背景的干扰,can base learning models on the presence of parts of the object instead of focusing on the whole object
random erasing,can be viewed as adding noise to the image
结合起来样本更加丰富
5 Experiments
5.1 Datasets and Metrics
1)数据集
分类
CIFAR-10
CIFAR-100
Fashion-MNIST检测
PASCAL VOC 2007ReID
Market-1501
DukeMTMC-reID
CUHK03
2)评价指标
分类,top-1 error rates,“mean std” based on 5 runs
检测,mAP
ReID,rank-1,mAP
5.2 Experiments
5.2.1 Image Classification
1)Classification accuracy on different datasets
p = 0.5 p = 0.5 p=0.5, s l = 0.02 s_l = 0.02 sl=0.02, s h = 0.4 s_h = 0.4 sh=0.4, and r 1 = 1 / r 2 = 0.3 r1 =1/r2= 0.3 r1=1/r2=0.3
2)The impact of hyper-parameters
fix s l s_l sl to 0.02, r 1 = 1 / r 2 r1 = 1/r2 r1=1/r2 and evaluate p p p, s h s_h sh, and r 1 r1 r1
We set p = 0.5 p = 0.5 p=0.5, s h = 0.4 s_h = 0.4 sh=0.4 and r 1 = 0.3 r1 = 0.3 r1=0.3 as the base setting. When evaluating one of the parameters, we fixed the other two parameters
都比 Baseline(没有 random erasing) 效果要更好!
3)Four types of random values for erasing

随机填充值和填充 ImageNet 的 mean 效果差不多[125, 122,114](算是明白 114 的出处了,原来那个灰白颜色是 ImageNet 的均值),优于填 0 和 255
4)Comparison with Dropout and random noise
random erasing 更胜一筹
5)Comparing with data augmentation methods
单独的话,random cropping > random flipping > random erasing
三合一 1+1+1>1,猛
6)Robustness to occlusion
手动遮挡,测试下效果 random erasing 的效果
We randomly select a region of area and fill it with random values. aspect ratio [0.3, 3.33]
秀
5.2.2 Object Detection

秀
5.2.3 Person Reidentification

秀
SOTA 配合 re-ranking,结果更进一层


6 Conclusion(own) / Future work
arXiv-2017-11-16 挂出来的论文中了 2020 AAAI,消失的时间
Deep learning: Dropout, DropConnect
12中主要的Dropout方法:如何应用于DNNs,CNNs,RNNs中的数学和可视化解释
- 《A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection》(CVPR-2017)
通过训练(对特征图进行遮挡和仿射变换)提升检测网络对遮挡、形变物体的识别精度
- 《PatchShuffle Regularization》(arXiv-2017)
边栏推荐
- 程序的执行
- 解决万恶的open failed: ENOENT (No such file or directory)/(Operation not permitted)
- [introduction to information retrieval] Chapter 1 Boolean retrieval
- Typeerror in allenlp: object of type tensor is not JSON serializable error
- [Bert, gpt+kg research] collection of papers on the integration of Pretrain model with knowledge
- SSM student achievement information management system
- Open failed: enoent (no such file or directory) / (operation not permitted)
- Calculate the total in the tree structure data in PHP
- SSM second hand trading website
- Faster-ILOD、maskrcnn_benchmark安装过程及遇到问题
猜你喜欢

半监督之mixmatch

ERNIE1.0 与 ERNIE2.0 论文解读

Sparksql data skew

Implementation of yolov5 single image detection based on onnxruntime
![[introduction to information retrieval] Chapter 1 Boolean retrieval](/img/78/df4bcefd3307d7cdd25a9ee345f244.png)
[introduction to information retrieval] Chapter 1 Boolean retrieval

Machine learning theory learning: perceptron

【AutoAugment】《AutoAugment:Learning Augmentation Policies from Data》

MMDetection安装问题

【Mixup】《Mixup:Beyond Empirical Risk Minimization》
![[in depth learning series (8)]: principles of transform and actual combat](/img/2e/89920de2273b6f1bc3b21a19c2ecbe.png)
[in depth learning series (8)]: principles of transform and actual combat
随机推荐
《Handwritten Mathematical Expression Recognition with Bidirectionally Trained Transformer》论文翻译
[tricks] whiteningbert: an easy unsupervised sentence embedding approach
基于onnxruntime的YOLOv5单张图片检测实现
深度学习分类优化实战
latex公式正体和斜体
ModuleNotFoundError: No module named ‘pytest‘
Deep learning classification Optimization Practice
Use matlab to realize: chord cut method, dichotomy, CG method, find zero point and solve equation
【深度学习系列(八)】:Transoform原理及实战之原理篇
Two dimensional array de duplication in PHP
程序的内存模型
Typeerror in allenlp: object of type tensor is not JSON serializable error
Conda 创建,复制,分享虚拟环境
Point cloud data understanding (step 3 of pointnet Implementation)
图片数据爬取工具Image-Downloader的安装和使用
conda常用命令
生成模型与判别模型的区别与理解
第一个快应用(quickapp)demo
ERNIE1.0 与 ERNIE2.0 论文解读
Module not found: Error: Can't resolve './$$_gendir/app/app.module.ngfactory'




