Local Magnification for Data and Feature Augmentation

11/15/2022
by   Kun He, et al.
0

In recent years, many data augmentation techniques have been proposed to increase the diversity of input data and reduce the risk of overfitting on deep neural networks. In this work, we propose an easy-to-implement and model-free data augmentation method called Local Magnification (LOMA). Different from other geometric data augmentation methods that perform global transformations on images, LOMA generates additional training data by randomly magnifying a local area of the image. This local magnification results in geometric changes that significantly broaden the range of augmentations while maintaining the recognizability of objects. Moreover, we extend the idea of LOMA and random cropping to the feature space to augment the feature map, which further boosts the classification accuracy considerably. Experiments show that our proposed LOMA, though straightforward, can be combined with standard data augmentation to significantly improve the performance on image classification and object detection. And further combination with our feature augmentation techniques, termed LOMA_IF FO, can continue to strengthen the model and outperform advanced intensity transformation methods for data augmentation.

READ FULL TEXT

page 2

page 5

page 6

research
06/22/2021

Data Augmentation for Opcode Sequence Based Malware Detection

Data augmentation has been successfully used in many areas of deep-learn...
research
07/22/2021

Geometric Data Augmentation Based on Feature Map Ensemble

Deep convolutional networks have become the mainstream in computer visio...
research
06/12/2023

Rotational augmentation techniques: a new perspective on ensemble learning for image classification

The popularity of data augmentation techniques in machine learning has i...
research
06/18/2018

HitNet: a neural network with capsules embedded in a Hit-or-Miss layer, extended with hybrid data augmentation and ghost capsules

Neural networks designed for the task of classification have become a co...
research
12/29/2022

Learning to mask: Towards generalized face forgery detection

Generalizability to unseen forgery types is crucial for face forgery det...
research
04/07/2022

TorMentor: Deterministic dynamic-path, data augmentations with fractals

We propose the use of fractals as a means of efficient data augmentation...
research
12/02/2020

A Self-Supervised Feature Map Augmentation (FMA) Loss and Combined Augmentations Finetuning to Efficiently Improve the Robustness of CNNs

Deep neural networks are often not robust to semantically-irrelevant cha...

Please sign up or login with your details

Forgot password? Click here to reset