arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.03394cs.CV
收藏

用于弱监督语义分割的注意力图背景噪声降低

Background Noise Reduction of Attention Map for Weakly Supervised Semantic Segmentation

Tokushima University(德岛大学)

Izumi Fujimori, Masaki Oono, Masami Shishibori

AI总结:

针对TransCAM注意力权重中的背景噪声问题,本文提出降噪方法以提升伪标签质量,并在PASCAL VOC 2012和MS COCO 2014上取得优于TransCAM的分割结果。

摘要:

在仅使用图像级类别标签的弱监督语义分割(WSSS)中,基于CNN的类激活图(CAM)存在一个问题,即它们倾向于激活物体最具判别性的局部区域。另一方面,基于Transformers的方法能够学习全局特征,但会受到背景噪声污染问题的影响。本文聚焦于解决现有基于Conformer、被称为TransCAM的WSSS方法中注意力权重的背景噪声问题。所提出的方法成功降低了背景噪声,从而提高了伪标签的准确性。实验结果表明,本文模型在PASCAL VOC 2012验证数据上取得70.5%的分割性能,在测试数据上取得71.1%,在MS COCO 2014数据上取得45.9%,在分割性能方面优于TransCAM。

英文摘要:

In weakly-supervised semantic segmentation (WSSS) using only image-level class labels, a problem with CNN-based Class Activation Maps (CAM) is that they tend to activate the most discriminative local regions of objects. On the other hand, methods based on Transformers learn global features but suffer from the issue of background noise contamination. This paper focuses on addressing the issue of background noise in attention weights within the existing WSSS method based on Conformer, known as TransCAM. The proposed method successfully reduces background noise, leading to improved accuracy of pseudo labels. Experimental results demonstrate that our model achieves segmentation performance of 70.5% on the PASCAL VOC 2012 validation data, 71.1% on the test data, and 45.9% on MS COCO 2014 data, outperforming TransCAM in terms of segmentation performance.

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。

↑