arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1903.04586cs.CV
收藏

使用深度图像表示生成超像素

Generating superpixels using deep image representations

Center for Processing Speech and Images(语音和图像处理中心);KU Leuven(比利时鲁汶大学);Dept. ESAT(ESAT系(ESAT系));ESAT(ESAT(比利时鲁汶大学的一个系/实验室));KU Leuven, Belgium(比利时鲁汶大学)

Thomas Verelst, Matthew Blaschko, Maxim Berman

摘要:

超像素算法是计算机视觉算法(例如分割、对象跟踪和定位)的常见预处理步骤。许多超像素方法仅依赖于用于分割的颜色特征,限制了低对比度区域中的性能以及对对象边界具有广泛外观变化的红外或医学图像的适用性。我们研究了SLIC超像素算法中包含的深度图像特征,以利用更高级别的图像表示。此外,我们设计了一个可训练的超像素算法,产生一个中间域特定的图像表示,可以应用于不同的任务。将基于聚类的超像素算法转化为逐像素分类任务,并从语义分割数据集导出超像素训练数据。我们的结果表明,这种方法能够始终如一地提高超像素质量。

英文摘要:

Superpixel algorithms are a common pre-processing step for computer vision algorithms such as segmentation, object tracking and localization. Many superpixel methods only rely on colors features for segmentation, limiting performance in low-contrast regions and applicability to infrared or medical images where object boundaries have wide appearance variability. We study the inclusion of deep image features in the SLIC superpixel algorithm to exploit higher-level image representations. In addition, we devise a trainable superpixel algorithm, yielding an intermediate domain-specific image representation that can be applied to different tasks. A clustering-based superpixel algorithm is transformed into a pixel-wise classification task and superpixel training data is derived from semantic segmentation datasets. Our results demonstrate that this approach is able to improve superpixel quality consistently.

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。

↑