Exploring Image Specific Structured Loss for Image Annotation with Incomplete Labelling
Exploring Image Specific Structured Loss for Image Annotation with Incomplete Labelling
复制标题
DOI：
10.1007/978-3-319-16865-4_46
复制
发表时间：
2014-11
期刊：
影响因子：
--
通讯作者：
Xing Xu;Atsushi Shimada;R. Taniguchi
中科院分区：
文献类型：
--
作者：
Xing Xu;Atsushi Shimada;R. Taniguchi
在本文中，我们解决的问题与不完全标记的图像标注，在每个训练图像中的多个对象没有完全标记。传统的one-versus-all SVM（OVA-SVM）在完全标记上表现得相当好，但在不完全设置下会急剧衰减。最近，结构化学习方法OVA-SSVM被提出来提高OVA-SVM的性能，通过建模标签的结构化关联，并在不完全设置下显示出效率。OVA-SSVM假设每个训练样本包含单个标签，并采用分类风格的损失度量，只要预测的标签之一是正确的，则整体预测应该被认为是正确的。但是，这可能不适合多标签注释任务。在本文中，我们将OVA-SSVM方法扩展到多标签的情况下，并设计了一种新的图像特定的结构化损失度量，以考虑依赖于图像标签关联的预测标签之间的依赖关系。然后，我们开发了一个有效的优化算法来学习模型参数。最后，我们提出了广泛的实证结果在两个基准数据集与不同程度的不完整性，并表明，该方法优于OVA-SSVM，并实现竞争力的性能相比，其他国家的最先进的方法，也是专为不完整的标签的问题。
In this paper, we address the problem of image annotation withincomplete labelling, where the multiple objects in each training image are not fully labeled. The conventional one-versus-all SVM (OVA-SVM) that performs fairly well on full labelling decays drastically under the incomplete setting. Recently, structured learning method termed OVA-SSVM is proposed to boost the performance of OVA-SVM by modeling the structured associations of labels and show efficiency under incomplete setting. The OVA-SSVM assumes that each training sample includes a single label and adopts an loss measure of classification style that as long as one of the predicted label is correct, the overall prediction should be considered correct. However, this may not be appropriate for the multi-label annotation task. In this paper, we extend the OVA-SSVM method to the multi-label situation and design a novel image specific structured loss measure to account for the dependencies between predicted labels relying on the image-label associations. Then we develop an efficient optimization algorithm to learn the model parameters. Finally, we present extensive empirical results on two benchmark datasets with various degree of incompletion, and show that proposed method outperforms OVA-SSVM and achieves competitive performance compared with other state-of-the-art methods which are also designed for the issue of incomplete labelling.