检索
E-mail
RSS
作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2010, Vol. 36 ›› Issue (06): 198-200.doi: 10.3969/j.issn.1000-3428.2010.06.067

• 人工智能及识别技术 • 上一篇    下一篇

基于核密度估计的图像自动标注方法

周 宁，薛向阳   

  1. (复旦大学计算机科学与技术学院，上海 200433)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2010-03-20 发布日期:2010-03-20

Image Automatic Annotation Method Based on Kernel Density Estimation

ZHOU Ning, XUE Xiang-yang   

  1. (School of Computer Science and Technology, Fudan University, Shanghai 200433)
  • Received:1900-01-01 Revised:1900-01-01 Online:2010-03-20 Published:2010-03-20

PDF

2379

被引次数

3 | 2

摘要： 提出一种基于概率模型的图像自动语义标注方法，将图片自动标注看作一个多类分类问题，通过无参数的核密度估计，实现用含有共同标注词的图片组估计视觉特征和相应标注词之间关系的机制。选取表达能力较好的基于CPAM的视觉特征，无须对图像进行语义分割处理，有效提高核密度估计的效率。在基准数据集上进行实验，结果表明，该模型能够获得比当前其他相关方法更好的标注性能。

关键词: 图像自动标注, 多类分类器, 核密度估计

Abstract: A novel method for automatic image annotation is presented. The new model is based on a probabilistic formulation which poses annotation as a multi-class classification problem. It tries to estimate the correlation between visual features and semantic labels by using the groups of images that share the same associated labels through kernel density estimation. In addition, CPAM-based visual features are introduced to improve the efficiency of kernel density estimation without requiring prior image semantic segmentation. Experiments on the benchmark data set show this model achieves higher accuracy than the previously published results.

Key words: image automatic annotation, multi-class classifier, kernel density estimation

中图分类号: