Attention-based Dropout Layer for Weakly Supervised Object Localization

TL;DR

This paper introduces an Attention-based Dropout Layer (ADL) that leverages self-attention to improve weakly supervised object localization by capturing entire objects rather than just discriminative parts, achieving state-of-the-art results.

Contribution

The paper proposes a novel ADL that enhances WSOL by hiding discriminative parts and highlighting informative regions, improving accuracy and efficiency.

Findings

Achieved new state-of-the-art localization accuracy on CUB-200-2011.

ADL improves recognition power by focusing on entire objects.

Method is more parameter- and computation-efficient than existing techniques.

Abstract

Weakly Supervised Object Localization (WSOL) techniques learn the object location only using image-level labels, without location annotations. A common limitation for these techniques is that they cover only the most discriminative part of the object, not the entire object. To address this problem, we propose an Attention-based Dropout Layer (ADL), which utilizes the self-attention mechanism to process the feature maps of the model. The proposed method is composed of two key components: 1) hiding the most discriminative part from the model for capturing the integral extent of object, and 2) highlighting the informative region for improving the recognition power of the model. Based on extensive experiments, we demonstrate that the proposed method is effective to improve the accuracy of WSOL, achieving a new state-of-the-art localization accuracy in CUB-200-2011 dataset. We also show that the proposed method is much more efficient in terms of both parameter and computation overheads than existing techniques.