SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularization

06/02/2020
by   A. F. M. Shahab Uddin, et al.
0

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 the random selection of the patch may not necessarily represent any 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 that leads the model to learn more appropriate feature representation. SaliencyMix achieves a new state-of-the-art top-1 error of 20.09 classification using ResNet-101 architecture and also improves the model robustness against adversarial perturbations. Furthermore, SaliencyMix trained model helps to improve the object detection performance.

READ FULL TEXT

page 4

page 8

research
12/21/2020

ResizeMix: Mixing Data with Preserved Object Information and True Labels

Data augmentation is a powerful technique to increase the diversity of d...
research
05/13/2019

CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features

Regional dropout strategies have been proposed to enhance the performanc...
research
12/16/2021

Saliency Grafting: Innocuous Attribution-Guided Mixup with Calibrated Label Mixing

The Mixup scheme suggests mixing a pair of samples to create an augmente...
research
04/10/2023

Use the Detection Transformer as a Data Augmenter

Detection Transformer (DETR) is a Transformer architecture based object ...
research
06/16/2021

Evolving Image Compositions for Feature Representation Learning

Convolutional neural networks for visual recognition require large amoun...
research
06/06/2019

Improving Robustness Without Sacrificing Accuracy with Patch Gaussian Augmentation

Deploying machine learning systems in the real world requires both high ...
research
03/29/2020

Attentive CutMix: An Enhanced Data Augmentation Approach for Deep Learning Based Image Classification

Convolutional neural networks (CNN) are capable of learning robust repre...

Please sign up or login with your details

Forgot password? Click here to reset