Optimal Tag Sets for Automatic Image Annotation
Optimal Tag Sets for Automatic Image Annotation
复制标题
DOI：
10.5244/c.25.1
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
S. Moran;V. Lavrenko
中科院分区：
文献类型：
--
作者：
S. Moran;V. Lavrenko
在本文中，我们介绍了一种新的形式的连续相关性模型（BSCRM），捕捉标签之间的相关性，在一个正式的和一致的方式。我们应用一个波束搜索算法，找到一个接近最佳的一组相互关联的标签的图像在一个时间内，是线性的搜索树的深度。我们进行检查的模型性能下不同的内核的图像特征分布的表示，并提出了一种方法，适应内核的数据集。与原始CRM模型相比，具有Minkowski内核的BS-CRM显着提高了42%的召回率和38%的精度，并且在标准Corel 5 k数据集上优于最近的基线。
In this paper we introduce a new form of the Continuous Relevance Model (the BSCRM) that captures the correlation between tags in a formal and consistent manner. We apply a beam search algorithm to find a near optimal set of mutually correlated tags for an image in a time that is linear in the depth of the search tree. We conduct an examination of the model performance under different kernels for the representation of the image feature distributions and suggest a method of adapting the kernel to the dataset. BS-CRM with a Minkowski kernel is found to significantly increase recall by 42% and precision by 38% over the original CRM model and outperforms more recent baselines on the standard Corel 5k dataset.