Self-supervised Learning with Local Attention-Aware Feature

TL;DR

This paper introduces a self-supervised learning method that effectively generates global and local attention-aware visual features, outperforming previous methods and even surpassing fully-supervised learning on certain datasets.

Contribution

The paper presents a novel self-supervised approach that learns attention-aware features by differentiating image transformations and patches, achieving superior performance on multiple datasets.

Findings

Outperforms previous best on Tiny-ImageNet by 1.03%

Outperforms previous best on STL-10 by 2.32%

Surpasses fully-supervised methods on STL-10

Abstract

In this work, we propose a novel methodology for self-supervised learning for generating global and local attention-aware visual features. Our approach is based on training a model to differentiate between specific image transformations of an input sample and the patched images. Utilizing this approach, the proposed method is able to outperform the previous best competitor by 1.03% on the Tiny-ImageNet dataset and by 2.32% on the STL-10 dataset. Furthermore, our approach outperforms the fully-supervised learning method on the STL-10 dataset. Experimental results and visualizations show the capability of successfully learning global and local attention-aware visual representations.