Hide-and-Seek: A Data Augmentation Technique for Weakly-Supervised Localization and Beyond

TL;DR

Hide-and-Seek is a versatile data augmentation method that improves weakly-supervised object localization and other visual recognition tasks by randomly hiding patches in training images to encourage models to focus on diverse relevant features.

Contribution

The paper introduces a novel data augmentation technique called Hide-and-Seek that enhances model performance across multiple visual recognition tasks, especially weakly-supervised localization.

Findings

Improves object localization accuracy in weakly-supervised settings

Enhances performance in various recognition tasks like classification and segmentation

Works with any network without requiring changes during testing

Abstract

We propose 'Hide-and-Seek' a general purpose data augmentation technique, which is complementary to existing data augmentation techniques and is beneficial for various visual recognition tasks. The key idea is to hide patches in a training image randomly, in order to force the network to seek other relevant content when the most discriminative content is hidden. Our approach only needs to modify the input image and can work with any network to improve its performance. During testing, it does not need to hide any patches. The main advantage of Hide-and-Seek over existing data augmentation techniques is its ability to improve object localization accuracy in the weakly-supervised setting, and we therefore use this task to motivate the approach. However, Hide-and-Seek is not tied only to the image localization task, and can generalize to other forms of visual input like videos, as well as other recognition tasks like image classification, temporal action localization, semantic segmentation, emotion recognition, age/gender estimation, and person re-identification. We perform extensive experiments to showcase the advantage of Hide-and-Seek on these various visual recognition problems.