Log In Sign Up

Evolving Image Compositions for Feature Representation Learning

by   Paola Cascante-Bonilla, et al.

Convolutional neural networks for visual recognition require large amounts of training samples and usually benefit from data augmentation. This paper proposes PatchMix, a data augmentation method that creates new samples by composing patches from pairs of images in a grid-like pattern. These new samples' ground truth labels are set as proportional to the number of patches from each image. We then add a set of additional losses at the patch-level to regularize and to encourage good representations at both the patch and image levels. A ResNet-50 model trained on ImageNet using PatchMix exhibits superior transfer learning capabilities across a wide array of benchmarks. Although PatchMix can rely on random pairings and random grid-like patterns for mixing, we explore evolutionary search as a guiding strategy to discover optimal grid-like patterns and image pairing jointly. For this purpose, we conceive a fitness function that bypasses the need to re-train a model to evaluate each choice. In this way, PatchMix outperforms a base model on CIFAR-10 (+1.91), CIFAR-100 (+5.31), Tiny Imagenet (+3.52), and ImageNet (+1.16) by significant margins, also outperforming previous state-of-the-art pairwise augmentation strategies.


page 1

page 2

page 7

page 12


Data Augmentation using Random Image Cropping and Patching for Deep CNNs

Deep convolutional neural networks (CNNs) have achieved remarkable resul...

CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features

Regional dropout strategies have been proposed to enhance the performanc...

SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularization

Advanced data augmentation strategies have widely been studied to improv...

Image Classification on Small Datasets via Masked Feature Mixing

Deep convolutional neural networks require large amounts of labeled data...

Unsupervised feature learning by augmenting single images

When deep learning is applied to visual object recognition, data augment...

Radioactive data: tracing through training

We want to detect whether a particular image dataset has been used to tr...

A Data Augmentation Method by Mixing Up Negative Candidate Answers for Solving Raven's Progressive Matrices

Raven's Progressive Matrices (RPMs) are frequently-used in testing human...