Improved Techniques For Weakly-Supervised Object Localization

TL;DR

This paper introduces enhanced data augmentation and learning techniques for weakly-supervised object localization, significantly improving accuracy by addressing the focus on only the most discriminative object parts.

Contribution

It presents novel data augmentation and CNN optimization methods that outperform existing techniques in weakly-supervised object localization.

Findings

Top-1 localization accuracy improved by up to 37.3%

Effective augmentation for less discriminative parts

Validated through extensive qualitative and quantitative experiments

Abstract

We propose an improved technique for weakly-supervised object localization. Conventional methods have a limitation that they focus only on most discriminative parts of the target objects. The recent study addressed this issue and resolved this limitation by augmenting the training data for less discriminative parts. To this end, we employ an effective data augmentation for improving the accuracy of the object localization. In addition, we introduce improved learning techniques by optimizing Convolutional Neural Networks (CNN) based on the state-of-the-art model. Based on extensive experiments, we evaluate the effectiveness of the proposed approach both qualitatively and quantitatively. Especially, we observe that our method improves the Top-1 localization accuracy by 21.4 - 37.3% depending on configurations, compared to the current state-of-the-art technique of the weakly-supervised object localization.