SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better RegularizationDownload PDF

Published: 12 Jan 2021, Last Modified: 05 May 2023ICLR 2021 PosterReaders: Everyone
Keywords: SaliencyMix, Saliency Guided Data Augmentation, Data Augmentation, Regularization
Abstract: Advanced data augmentation strategies have widely been studied to improve the generalization ability of deep learning models. Regional dropout is one of the popular solutions that guides the model to focus on less discriminative parts by randomly removing image regions, resulting in improved regularization. However, such information removal is undesirable. On the other hand, recent strategies suggest to randomly cut and mix patches and their labels among training images, to enjoy the advantages of regional dropout without having any pointless pixel in the augmented images. We argue that such random selection strategies of the patches may not necessarily represent sufficient information about the corresponding object and thereby mixing the labels according to that uninformative patch enables the model to learn unexpected feature representation. Therefore, we propose SaliencyMix that carefully selects a representative image patch with the help of a saliency map and mixes this indicative patch with the target image, thus leading the model to learn more appropriate feature representation. SaliencyMix achieves the best known top-1 error of $21.26\%$ and $20.09\%$ for ResNet-50 and ResNet-101 architectures on ImageNet classification, respectively, and also improves the model robustness against adversarial perturbations. Furthermore, models that are trained with SaliencyMix, help to improve the object detection performance. Source code is available at \url{}.
One-sentence Summary: The proposed method carefully selects a representative image patch with the help of a saliency map and mixes that indicative patch with the target image that leads the model to learn more appropriate feature representation
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
Supplementary Material: zip
Code: [![github](/images/github_icon.svg) SaliencyMix/SaliencyMix]( + [![Papers with Code](/images/pwc_icon.svg) 1 community implementation](
Data: [CIFAR-10](, [CIFAR-100](
16 Replies