Abstract
ROI extraction is an active but challenging task in remote sensing because of the complicated landform, the complex boundaries and the requirement of annotations. Weakly supervised learning (WSL) aims at learning a mapping from input image to pixel-wise prediction under image-wise labels, which can dramatically decrease the labor cost. However, due to the imprecision of labels, the accuracy and time consumption of WSL methods are relatively unsatisfactory. In this paper, we propose a two-step ROI extraction based on contractive learning. Firstly, we present to integrate multiscale Grad-CAM to obtain pseudo pixelwise annotations with well boundaries. Then, to reduce the compact of misjudgments in pseudo annotations, we construct a contrastive learning strategy to encourage the features inside ROI as close as possible and separate background features from foreground features. Comprehensive experiments demonstrate the superiority of our proposal. Code is available at this https URL
Abstract (translated)
ROI提取在遥感中是一项具有活力但具有挑战性的任务,因为复杂的地形、复杂的边界和标注要求。弱监督学习(WSL)旨在学习从输入图像到像素级别的预测映射,该映射可以戏剧性地降低劳动力成本。然而,由于标签的不准确性,WSL方法的精度和时间消耗相对不太满意。在本文中,我们提出了基于收缩学习的两步ROI提取方法。首先,我们将融合多尺度grad-CAM,以获得伪像素级别的标注,具有良好的边界。然后,为了减少伪标注中的误判密集,我们建立了对比学习策略,以鼓励 ROI内部特征尽可能接近,并分离背景特征和前景特征。综合实验证明了我们提议的优越性。代码可在这个https URL上获取。
URL
https://arxiv.org/abs/2305.05887