Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

GridMask Data Augmentation

About

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.

Pengguang Chen, Shu Liu, Hengshuang Zhao, Xingquan Wang, Jiaya Jia• 2020

Related benchmarks

TaskDatasetResultRank
Object DetectionCOCO 2017 (val)
AP42.6
2454
Texture ClassificationDTD
Accuracy25.3
108
ClassificationCaltech101
Accuracy51.6
34
Fine grained classificationPets
Accuracy37.6
22
Fine grained classificationCars
Accuracy28.4
21
Image ClassificationCIFAR100 Subset
Accuracy48.2
19
Image ClassificationOrganSMNIST
Accuracy78.9
19
Image ClassificationBreastMNIST
Accuracy66.8
19
Image ClassificationPathMNIST
Accuracy78.4
19
Fine-grained Image ClassificationFlowers (test)
Accuracy80.7
14
Showing 10 of 11 rows

Other info

Follow for update