Papers › GridMask Data Augmentation

GridMask Data Augmentation

13 Jan 2020arXiv:2001.04086archive 2025-07-28

Pengguang Chen, Shu Liu, Hengshuang Zhao, Xingquan Wang, Jiaya Jia

We propose a novel data augmentation method `GridMask' in this paper. It utilizes information removal to achieve state-of-the-art results in a variety of computer vision tasks. We analyze the requirement of information dropping. Then we show limitation of existing information dropping algorithms and propose our structured method, which is simple and yet very effective. It is based on the deletion of regions of the input image. Our extensive experiments show that our method outperforms the latest AutoAugment, which is way more computationally expensive due to the use of reinforcement learning to find the best policies. On the ImageNet dataset for recognition, COCO2017 object detection, and on Cityscapes dataset for semantic segmentation, our method all notably improves performance over baselines. The extensive experiments manifest the effectiveness and generality of the new method.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

akuxcw/GridMask officialmentioned in papermentioned on GitHubpytorch report
IlyaDobrynin/GridMixup mentioned on GitHubpytorch report
Jia-Research-Lab/GridMask mentioned on GitHubpytorch report
Kaushal28/Bengali-AI mentioned on GitHubpytorch report
ma7555/Augz mentioned on GitHub report
PaddlePaddle/PaddleClas paddleApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Data AugmentationObject DetectionReinforcement LearningReinforcement Learning (RL)Semantic Segmentationobject-detectionreinforcement-learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Introduced by this paper: GridMask

AutoAugmentGridMaskLSTMSigmoid ActivationTanh Activation

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections