Modeling continuous visual features for semantic image annotation and retrieval

Automatic image annotation has become an important and challenging problem due to the existence of semantic gap. In this paper, we firstly extend probabilistic latent semantic analysis (PLSA) to model continuous quantity. In addition, corresponding Expectation-Maximization (EM) algorithm is derived to determine the model parameters. Furthermore, in order to deal with the data of different modalities in terms of their characteristics, we present a semantic annotation model which employs continuous PLSA and standard PLSA to model visual features and textual words respectively. The model learns the correlation between these two modalities by an asymmetric learning approach and then it can predict semantic annotation precisely for unseen images. Finally, we compare our approach with several state-of-the-art approaches on the Corel5k and Corel30k datasets. The experiment results show that our approach performs more effectively and accurately.     

A probabilistic semantic model for image annotation and multi-modal image retrieval

This paper addresses automatic image annotation problem and its application to multi-modal image retrieval. The contribution of our work is three-fold. (1) We propose a probabilistic semantic model in which the visual features and the textual words are connected via a hidden layer which constitutes the semantic concepts to be discovered to explicitly exploit the synergy among the modalities. (2) The association of visual features and textual words is determined in a Bayesian framework such that the confidence of the association can be provided. (3) Extensive evaluation on a large-scale, visually and semantically diverse image collection crawled from Web is reported to evaluate the prototype system based on the model. In the proposed probabilistic model, a hidden concept layer which connects the visual feature and the word layer is discovered by fitting a generative model to the training image and annotation words through an Expectation-Maximization (EM) based iterative learning procedure. The evaluation of the prototype system on 17,000 images and 7736 automatically extracted annotation words from crawled Web pages for multi-modal image retrieval has indicated that the proposed semantic model and the developed Bayesian framework are superior to a state-of-the-art peer system in the literature.     

结合相关反馈日志与语义网络的图像标注

连接高层语义和低层视觉特征的图像语义标注技术能够很好地表示图像的语义,提出并实现了一种结合相关反馈日志与语义网络的图像标注方法。该方法以收集的用户相关反馈日志为基础获得图像的语义信息,通过计算图像间的语义相似度进行语义聚类并采用语义传播的方式实现图像的语义标注。实验结果表明,随着相关反馈日志库的不断扩充,图像库中越来越多的图像会在反馈的过程中得到标注且标注的准确率会随着反馈次数的增加而趋于稳定。     

混合生成式和判别式模型的图像自动标注

目的 由于图像检索中存在着低层特征和高层语义之间的“语义鸿沟”,图像自动标注成为当前的关键性问题.为缩减语义鸿沟,提出了一种混合生成式和判别式模型的图像自动标注方法.方法 在生成式学习阶段,采用连续的概率潜在语义分析模型对图像进行建模,可得到相应的模型参数和每幅图像的主题分布.将这个主题分布作为每幅图像的中间表示向量,那么图像自动标注的问题就转化为一个基于多标记学习的分类问题.在判别式学习阶段,使用构造集群分类器链的方法对图像的中间表示向量进行学习,在建立分类器链的同时也集成了标注关键词之间的上下文信息,因而能够取得更高的标注精度和更好的检索效果.结果 在两个基准数据集上进行的实验表明,本文方法在Corel5k数据集上的平均精度、平均召回率分别达到0.28和0.32,在IAPR-TC12数据集上则达到0.29和0.18,其性能优于大多数当前先进的图像自动标注方法.此外,从精度—召回率曲线上看,本文方法也优于几种典型的具有代表性的标注方法.结论 提出了一种基于混合学习策略的图像自动标注方法,集成了生成式模型和判别式模型各自的优点,并在图像语义检索的任务中表现出良好的有效性和鲁棒性.本文方法和技术不仅能应用于图像检索和识别的领域,经过适当的改进之后也能在跨媒体检索和数据挖掘领域发挥重要作用.     

A vision-based postproduction tool for footage logging, analysis, and annotation

We report a new method for logging and annotating video footage directed towards the needs of professional postproduction and archivist end users. SALSA—Semi-Automated Logging with Semantic Annotation—is a hybrid system that uses automated footage analysis for cut detection and camera movement classification, and a stenographic-like keyboard input system for the logging of higher-level semantic information. Output is presented both in standard printed log form, with the addition of mosaic visual representations of shots, and in a fully searchable database. Experimental comparisons of SALSA with conventional hand analysis show a significant increase in the logger’s speed with no reduction in accuracy or semantic detail.     

In depth analysis and evaluation of saliency-based color image indexing methods using wavelet salient features

Nowadays, most of the research works in the area of image retrieval try to build an image signature by considering the image as a whole. In this paper, we proposed an alternative based on the detection of some salient points in the image. For this purpose, we propose a new efficient salient point detector based on a wavelet transform. The efficiency of our detector lies in the representation of the wavelet coefficients by a zerotree data structure and by a saliency formulation that does not favor any direction. Thus, the detected salient points are located on sharp region boundaries whatever their direction. From the detected salient points, we build a color/texture signature by using jointly the well-known color correlogram extended to salient features and rotated wavelet filter responses. Experimental results conducted by adopting a global salient approach and a local salient approach show the effectiveness of the proposed scheme.     

混合词汇特征和LDA的语义相关度计算方法

文本语义相关度计算在自然语言处理、语义信息检索等方面起着重要作用,以Wikipedia为知识库,基于词汇特征的ESA（Explicit Semantic Analysis）因简单有效的特点在这些领域中受到学术界的广泛关注和应用。然而其语义相关度计算因为有大量冗余概念的参与变成了一种高维度、低效率的计算方式,同时也忽略了文本所属主题因素对语义相关度计算的作用。引入LDA（Latent Dirichlet Allocation）主题模型,对ESA返回的相关度较高的概念转换为模型的主题概率向量,从而达到降低维度和提高效率的目的;将JSD距离（Jensen-Shannon Divergence）替换余弦距离的测量方法,使得文本语义相关度计算更加合理和有效。最后对不同层次的数据集进行算法的测试评估,结果表明混合词汇特征和主题模型的语义相关度计算方法的皮尔逊相关系数比ESA和LDA分别高出3%和9%以上。     

A novel level set method for image segmentation by incorporating local statistical analysis and global similarity measurement

This paper presents a novel level set method for complex image segmentation, where the local statistical analysis and global similarity measurement are both incorporated into the construction of energy functional. The intensity statistical analysis is performed on local circular regions centered in each pixel so that the local energy term is constructed in a piecewise constant way. Meanwhile, the Bhattacharyya coefficient is utilized to measure the similarity between probability distribution functions for intensities inside and outside the evolving contour. The global energy term can be formulated by minimizing the Bhattacharyya coefficient. To avoid the time-consuming re-initialization step, the penalty energy term associated with a new double-well potential is constructed to maintain the signed distance property of level set function. The experiments and comparisons with four popular models on synthetic and real images have demonstrated that our method is efficient and robust for segmenting noisy images, images with intensity inhomogeneity, texture images and multiphase images.