IEEE Trans Pattern Anal Mach Intell. 2011 Jul;33(7):1281-94. doi: 10.1109/TPAMI.2010.204. Epub 2010 Nov 18.
The increasing availability of large quantities of user contributed images with labels has provided opportunities to develop automatic tools to tag images to facilitate image search and retrieval. In this paper, we present a novel hybrid probabilistic model (HPM) which integrates low-level image features and high-level user provided tags to automatically tag images. For images without any tags, HPM predicts new tags based solely on the low-level image features. For images with user provided tags, HPM jointly exploits both the image features and the tags in a unified probabilistic framework to recommend additional tags to label the images. The HPM framework makes use of the tag-image association matrix (TIAM). However, since the number of images is usually very large and user-provided tags are diverse, TIAM is very sparse, thus making it difficult to reliably estimate tag-to-tag co-occurrence probabilities. We developed a collaborative filtering method based on nonnegative matrix factorization (NMF) for tackling this data sparsity issue. Also, an L1 norm kernel method is used to estimate the correlations between image features and semantic concepts. The effectiveness of the proposed approach has been evaluated using three databases containing 5,000 images with 371 tags, 31,695 images with 5,587 tags, and 269,648 images with 5,018 tags, respectively.
大量用户贡献的带标签图像的可用性不断增加，为开发自动标记图像以方便图像搜索和检索的工具提供了机会。在本文中，我们提出了一种新颖的混合概率模型 (HPM)，该模型集成了低水平的图像特征和高水平的用户提供的标签，以自动标记图像。对于没有任何标签的图像，HPM 仅基于低水平的图像特征来预测新的标签。对于具有用户提供的标签的图像，HPM 联合利用图像特征和标签在统一的概率框架中，以推荐其他标签来标记图像。HPM 框架利用了标签-图像关联矩阵 (TIAM)。然而，由于图像的数量通常非常大，并且用户提供的标签是多样化的，TIAM 非常稀疏，因此难以可靠地估计标签之间的共现概率。我们开发了一种基于非负矩阵分解 (NMF) 的协同过滤方法来解决这个数据稀疏性问题。此外，还使用 L1 范数核方法来估计图像特征和语义概念之间的相关性。使用分别包含 5000 张图像和 371 个标签、31695 张图像和 5587 个标签以及 269648 张图像和 5018 个标签的三个数据库评估了所提出方法的有效性。
