利用对抗擦除与伪标签改进弱监督目标定位
Improving Weakly-Supervised Object Localization Using Adversarial Erasing and Pseudo Label
AI总结：
针对弱监督目标定位仅用图像级标签难以定位完整目标的问题，提出结合对抗擦除与伪标签的框架，通过新型损失和伪标签优化激活，在多个骨干网络和公开数据集上性能超越现有最优方法。
摘要：
弱监督学习方法因能降低神经网络训练中人工标注的工作量而受到广泛关注。本文研究了一种弱监督目标定位框架，其目标是仅利用图像及其图像级类别标签，训练出可同时预测目标类别及其位置的神经网络。该框架由共享特征提取器、分类器和定位器组成：定位器预测像素级类别概率，分类器则在图像级预测目标类别。由于图像级类别标签不足以训练定位器，弱监督目标定位方法通常难以准确定位完整的目标区域。为解决这一问题，所提方法引入对抗擦除和伪标签以提升定位精度。具体而言，本文设计了新型损失函数，利用对抗擦除的前景特征和对抗擦除的特征图，降低对最具判别性区域的依赖。此外，该方法采用伪标签抑制背景区域的激活值，同时提升前景区域的激活值。研究将所提方法应用于两种骨干网络（MobileNetV1和InceptionV3），并在三个公开数据集（ILSVRC-2012、CUB-200-2011和PASCAL VOC 2012）上进行评估。实验结果表明，该方法在所有评估指标上均优于此前的最优方法。
英文摘要：
Weakly-supervised learning approaches have gained significant attention due to their ability to reduce the effort required for human annotations in training neural networks. This paper investigates a framework for weakly-supervised object localization, which aims to train a neural network capable of predicting both the object class and its location using only images and their image-level class labels. The proposed framework consists of a shared feature extractor, a classifier, and a localizer. The localizer predicts pixel-level class probabilities, while the classifier predicts the object class at the image level. Since image-level class labels are insufficient for training the localizer, weakly-supervised object localization methods often encounter challenges in accurately localizing the entire object region. To address this issue, the proposed method incorporates adversarial erasing and pseudo labels to improve localization accuracy. Specifically, novel losses are designed to utilize adversarially erased foreground features and adversarially erased feature maps, reducing dependence on the most discriminative region. Additionally, the proposed method employs pseudo labels to suppress activation values in the background while increasing them in the foreground. The proposed method is applied to two backbone networks (MobileNetV1 and InceptionV3) and is evaluated on three publicly available datasets (ILSVRC-2012, CUB-200-2011, and PASCAL VOC 2012). The experimental results demonstrate that the proposed method outperforms previous state-of-the-art methods across all evaluated metrics.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
