Cutout is an image augmentation and regularization technique that randomly masks out square regions of input during training. and can be used to improve the robustness and overall performance of convolutional neural networks. The main motivation for cutout comes from the problem of object occlusion, which is commonly encountered in many computer vision tasks, such as object recognition, tracking, or human pose estimation. By generating new images which simulate occluded examples, we not only better prepare the model for encounters with occlusions in the real world, but the model also learns to take more of the image context into consideration when making decisions

Source: Improved Regularization of Convolutional Neural Networks with Cutout

Latest Papers

PAPER DATE
MaxDropout: Deep Neural Network Regularization Based on Maximum Output Values
| Claudio Filipi Goncalves do SantosDanilo ColomboMateus RoderJoão Paulo Papa
2020-07-27
SSN: Soft Shadow Network for Image Compositing
Yichen ShengJianming ZhangBedrich Benes
2020-07-16
Delving into the Adversarial Robustness on Face Recognition
| Xiao YangDingcheng YangYinpeng DongWenjian YuHang SuJun Zhu
2020-07-08
Differentiable Augmentation for Data-Efficient GAN Training
| Shengyu ZhaoZhijian LiuJi LinJun-Yan ZhuSong Han
2020-06-18
Angle-based Search Space Shrinking for Neural Architecture Search
| Yiming HuYuding LiangZichao GuoRuosi WanXiangyu ZhangYichen WeiQingyi GuJian Sun
2020-04-28
ResNeSt: Split-Attention Networks
| Hang ZhangChongruo WuZhongyue ZhangYi ZhuZhi ZhangHaibin LinYue SunTong HeJonas MuellerR. ManmathaMu LiAlexander Smola
2020-04-19
How Not to Give a FLOP: Combining Regularization and Pruning for Efficient Inference
| Tai VuEmily WenRoy Nehoran
2020-03-30
TResNet: High Performance GPU-Dedicated Architecture
| Tal RidnikHussam LawenAsaf NoyItamar FriedmanEmanuel Ben BaruchGilad Sharir
2020-03-30
Attentive CutMix: An Enhanced Data Augmentation Approach for Deep Learning Based Image Classification
Devesh WalawalkarZhiqiang ShenZechun LiuMarios Savvides
2020-03-29
Getting to 99% Accuracy in Interactive Segmentation
| Marco ForteBrian PriceScott CohenNing XuFrançois Pitié
2020-03-17
Can We Read Speech Beyond the Lips? Rethinking RoI Selection for Deep Visual Speech Recognition
Yuanhang ZhangShuang YangJingyun XiaoShiguang ShanXilin Chen
2020-03-06
Data augmentation with Mobius transformations
Sharon ZhouJiequan ZhangHang JiangTorbjorn LundhAndrew Y. Ng
2020-02-07
Compounding the Performance Improvements of Assembled Techniques in a Convolutional Neural Network
| Jungkyu LeeTaeryun WonTae Kwan LeeHyemin LeeGeonmo GuKiho Hong
2020-01-17
UNAS: Differentiable Architecture Search Meets Reinforcement Learning
Arash VahdatArun MallyaMing-Yu LiuJan Kautz
2019-12-16
Neural Predictor for Neural Architecture Search
| Wei WenHanxiao LiuHai LiYiran ChenGabriel BenderPieter-Jan Kindermans
2019-12-02
Adversarial Examples Improve Image Recognition
| Cihang XieMingxing TanBoqing GongJiang WangAlan YuilleQuoc V. Le
2019-11-21
RandAugment: Practical automated data augmentation with a reduced search space
| Ekin D. CubukBarret ZophJonathon ShlensQuoc V. Le
2019-09-30
Improved Image Augmentation for Convolutional Neural Networks by Copyout and CopyPairing
| Philip May
2019-09-01
SCARLET-NAS: Bridging the gap between Stability and Scalability in Weight-sharing Neural Architecture Search
| Xiangxiang ChuBo ZhangJixiang LiQingyuan LiRuijun Xu
2019-08-16
Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation
Raphael Gontijo LopesDong YinBen PooleJustin GilmerEkin D. Cubuk
2019-06-06
Semi-supervised semantic segmentation needs strong, varied perturbations
| Geoff FrenchSamuli LaineTimo AilaMichal MackiewiczGraham Finlayson
2019-06-05
Fast AutoAugment
| Sungbin LimIldoo KimTaesup KimChiheon KimSungwoong Kim
2019-05-01
MultiGrain: a unified image embedding for classes and instances
| Maxim BermanHervé JégouAndrea VedaldiIasonas KokkinosMatthijs Douze
2019-02-14
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
| Han CaiLigeng ZhuSong Han
2018-12-02
Data Augmentation using Random Image Cropping and Patching for Deep CNNs
| Ryo TakahashiTakashi MatsubaraKuniaki Uehara
2018-11-22
DropFilter: A Novel Regularization Method for Learning Convolutional Neural Networks
Hengyue PanHui JiangXin NiuYong Dou
2018-11-16
Improved Regularization of Convolutional Neural Networks with Cutout
| Terrance DeVriesGraham W. Taylor
2017-08-15

Components

COMPONENT TYPE
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign

Categories