arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.12459cs.CVcs.LGeess.IV
收藏

用于弱监督分割的重要性采样CAM

Importance Sampling CAMs for Weakly-Supervised Segmentation

Husqvarna Group(胡斯华纳集团);University of KwaZulu-Natal(夸祖鲁-纳塔尔大学)

Arvi Jonnarth, Michael Felsberg

AI总结:

针对弱监督分割中CAMs仅关注判别区域且预测轮廓模糊的问题,提出基于重要性采样的随机类别预测和特征相似性损失项,显著提升了轮廓精度并保持了区域相似性。

摘要:

分类网络可以通过类激活图(CAMs)来定位和分割图像中的物体。然而,在没有像素级标注的情况下,分类网络已知会(1)主要关注具有判别性的区域,以及(2)产生没有明确预测轮廓的扩散CAMs。在这项工作中,我们通过两项改进CAM学习的贡献来解决这两个问题。首先,我们引入基于CAMs诱导的逐类概率质量函数的重要性采样,以产生随机图像级类别预测。这使得CAMs能够在物体更大范围内激活。其次,我们提出了一个特征相似性损失项,旨在使预测轮廓与图像中的边缘相匹配。作为第三项贡献,我们在PASCAL VOC 2012基准数据集上进行了实验,证明这些修改在轮廓精度方面显著提高了性能,同时在区域相似性方面与当前最先进的方法相当。

英文摘要:

Classification networks can be used to localize and segment objects in images by means of class activation maps (CAMs). However, without pixel-level annotations, classification networks are known to (1) mainly focus on discriminative regions, and (2) to produce diffuse CAMs without well-defined prediction contours. In this work, we approach both problems with two contributions for improving CAM learning. First, we incorporate importance sampling based on the class-wise probability mass function induced by the CAMs to produce stochastic image-level class predictions. This results in CAMs which activate over a larger extent of objects. Second, we formulate a feature similarity loss term which aims to match the prediction contours with edges in the image. As a third contribution, we conduct experiments on the PASCAL VOC 2012 benchmark dataset to demonstrate that these modifications significantly increase the performance in terms of contour accuracy, while being comparable to current state-of-the-art methods in terms of region similarity.

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。

↑