基于潜子空间去噪的子空间学习图像分类方法

针对判别最小二乘回归(DLSR)对图像噪声鲁棒性不佳的问题,提出一种基于潜子空间去噪的子空间学习图像分类方法(DLSSL).该方法在架构上不同于现有基于回归的分类方法,其在视觉空间与标签空间中引入一个潜在子空间,将传统的图像分类框架改进为两步,即先降噪后分类.该方法先通过欠完备自编码将数据的高阶特征提取到潜在空间,再利用此高阶特征进行回归分类,同时辅以组核范数约束控制类内样本间距离.潜在子空间的引入为算法框架带来了更多灵活性,缓解了视觉空间与标签空间中数据维度与特性的差异,使得欠完备自编码可以有效地对数据进行降噪,提升了分类算法的鲁棒性.在人脸、生物指纹、物体和深度特征数据集上设计了多组对比实验,实验结果表明,算法对于图像中的噪声具有较强的鲁棒性,获得的投影矩阵具有良好的判别性,相比现有图像分类算法,性能更好、普适性更强,能有效地运用于各种图像分类任务.     

Feature Extraction Based on Maximum Nearest Subspace Margin Criterion

Based on the classification rule of sparse representation-based classification (SRC) and linear regression classification (LRC), we propose the maximum nearest subspace margin criterion for feature extraction. The proposed method can be seen as a preprocessing step of SRC and LRC. By maximizing the inter-class reconstruction error and minimizing the intra-class reconstruction error simultaneously, the proposed method significantly improves the performances of SRC and LRC. Compared with linear discriminant analysis, the proposed method avoids the small sample size problem and can extract more features. Moreover, we extend LRC to overcome the potential singular problem. The experimental results on the extended Yale B (YALE-B), AR, PolyU finger knuckle print and the CENPARMI handwritten numeral databases demonstrate the effectiveness of the proposed method.     

基于特征子空间学习的跨媒体检索方法

学习不同模态的多媒体数据在底层特征上的潜在关系,在降维得到的特征子空间中通过基于相似度传递的优化算法对图像和音频的聚类质量进行修正.相关反馈过程中设计了3种主动学习策略用以计算用户标注样本周围未标注样本的条件概率,从而在反馈样本有限的情况下提高跨媒体检索效率.实验结果表明该方法准确度量跨媒体的相关性,有效实现图像和音频数据之间的相互检索.     

基于子空间增量学习的视频中人脸图像检索

提出一种以电影视频中人脸图像为依据的视频检索方法.首先通过AdaBoost检测视频序列中的人脸图像,将检测到的人脸做标准化处理后投影到增量特征人脸子空间中,得到人脸图像的向量表述;然后应用单类支持向量机进行训练和分类,根据分类的结果动态地调整前面得到的最优分类超平面,实现对电影视频中特定演员的检索功能.由于不同镜头中同一人的人脸图像通常差别很大,该方法随时间序列动态地调整特征人脸空间,以适应人脸分布的变化.对电影《小花》、《Notting hill》等的实验表明,该方法在视频环境下可以较准确地检索出特定人像.     

图像检索中的主动学习及其可测量性

主动学习对于复杂、主观、使用少量训练实例的图像检索查询具有非常有效的作用。在图像检索中应用主动学习与支持向量机相结合的方法进行相关反馈,通过两者的互补来有效地提高图像检索的精度。对比了推理算法、简单主动算法以及角度多样性算法3种主动学习算法,并研究了最好的样本选择策略。还讨论了主动学习中概念复杂度的可测量性,并对未来的研究方向提出了建议。相信随着这些可测量性问题被重点提出,主动学习的成果可以被广泛应用。     

图像检索中的主动学习及其可测量性

主动学习对于复杂、主观、使用少量训练实例的图像检索查询具有非常有效的作用。在图像检索中应用主动学习与支持向量机相结合的方法进行相关反馈，通过两者的互补来有效地提高图像检索的精度。对比了推理算法、简单主动算法以及角度多样性算法3种主动学习算法，并研究了最好的样本选择策略。还讨论了主动学习中概念复杂度的可测量性，并对未来的研究方向提出了建议。相信随着这些可测量性问题被重点提出，主动学习的成果可以被广泛应用。     

Learning Multifunctional Binary Codes for Personalized Image Retrieval

International Journal of Computer Vision - Due to the highly complex semantic information of images, even with the same query image, the expected content-based image retrieval results could be very...     

图像检索中的用户兴趣学习

提出一个基于Bayesian学习的用户兴趣模型，用随机变量来刻画用户在图像检索过程中的个性倾向。重点分析了在随机变量的分布函数是动态变化的一族线性函数时，模型对检索效果的影响。实验结果显示，利用线性族的动态分布函数比固定正态分布函数以及均匀分布函数会获得更好的检索效果。     

基于共同子空间分类学习的跨媒体检索研究

不同媒体数据间由于存在严重的异构鸿沟和语义鸿沟,而不能直接计算它们之间的语义相似度,从而影响了跨媒体检索的实现和效果.当前提出的共同子空间学习虽能实现跨媒体语义关联和检索,但多采用一般的特征提取技术,且在语义匹配时的分类效果较差,不能有效实现跨媒体数据的高层语义关联计算,影响了检索效果.对此,提出Stacking-DS...     

区域图像检索中的AdaBoost反馈学习算法

提出了一种基于AdaBoost相关反馈的区域图像检索方法.结合图分割的图像分割算法和多区域匹配方法,利用用户的反馈信息对AdaBoost弱分类器的反复训练,得到一个具有较小错误率的强分类器.将其应用到区域图像检索中从而返回更加精确的查询结果.实验表明(图像数据库大小为10000),基于AdaBoost相关反馈的区域图像检索方法有更好的检索查准率和密集度,优于单一区域的图像比较算法和多区域比较算法.     

Region-based Image Retrieval Using Probabilistic Feature Relevance Learning

Region-Based Image Retrieval (RBIR), a specialisation of content-based image retrieval, is a promising and important research area. RBIR usually requires good segmentation, which is often difficult to achieve in practice for several reasons, such as varying environmental conditions and occlusion. It is, therefore, imperative to develop effective mechanisms for interactive, region-based visual query in order to provide confident retrieval performance. In this paper, we present a novel RBIR system, Finding Region In the Pictures (FRIP), that uses human-centric relevance feedback to create similarity metric on-the-fly in order to overcome some of the limitations associated with RBIR systems. We use features such as colour, texture, normalised area, shape and location, extracted from each region of a segmented image, to represent image content. For each given query, we estimate local feature relevance using probabilistic relevance model, from which to create a flexible metric that is highly adaptive to query location. As a result, local data densities can be sufficiently exploited, whereby rapid performance improvement can be achieved. The efficacy of our method is validated and compared against other competing techniques using real world image data.     

A Framework for Robust Subspace Learning

Many computer vision, signal processing and statistical problems can be posed as problems of learning low dimensional linear or multi-linear models. These models have been widely used for the representation of shape, appearance, motion, etc., in computer vision applications. Methods for learning linear models can be seen as a special case of subspace fitting. One draw-back of previous learning methods is that they are based on least squares estimation techniques and hence fail to account for outliers which are common in realistic training sets. We review previous approaches for making linear learning methods robust to outliers and present a new method that uses an intra-sample outlier process to account for pixel outliers. We develop the theory of Robust Subspace Learning (RSL) for linear models within a continuous optimization framework based on robust M-estimation. The framework applies to a variety of linear learning problems in computer vision including eigen-analysis and structure from motion. Several synthetic and natural examples are used to develop and illustrate the theory and applications of robust subspace learning in computer vision.     

Image Annotation and Retrieval for Weakly Labeled Images Using Conceptual Learning

One of the most promising new technologies for widespread application is image annotation and retrieval. Nevertheless, this task is very difficult to accomplish as target images differ significantly in appearance and belong to a wide variety of categories. In this paper, we propose a new image annotation and retrieval method for miscellaneous weakly labeled images, by combining higher-order local auto-correlation (HLAC) features and a framework of probabilistic canonical correlation analysis. The distance between images can be defined in the intrinsic space for annotation using conceptual learning of images and their labels. Because this intrinsic space is highly compressed compared to the image feature space, our method achieves both faster and more accurate image annotation and retrieval. The HLAC features are powerful global features with additive and position invariant properties. These properties work well with images, which have an arbitrary number of objects at arbitrary locations. The proposed method is shown to outperform existing methods using a standard benchmark dataset.     

基于权重学习的图像最大权对集胚配模型

在图匹配模型中权重的设置对匹配性能有很大影响,但直接计算的权重往往不符合匹配图像的实际情况。为此,参照二次分配问题的图匹配学习思想,给出一阶和二阶最大权对集模型的权重学习计算方法。一阶最大权对集模型直接采用图像特征点作为图的顶点,而二阶最大权对集模型则采用某些特征点之间的连接边作为顶点,2个模型都可以通过Kuhn—Munkras算法求解。一阶最大权对集模型在本质上等价于二次分配问题的线性情况。在CMUHouse数据库上的图像匹配实验结果表明,二阶最大权对集模型优于一阶最大权对集模型,且两者在学习计算时的性能也优于直接计算的情况。     

Maximum Margin Clustering Made Practical

Motivated by the success of large margin methods in supervised learning, maximum margin clustering (MMC) is a recent approach that aims at extending large margin methods to unsupervised learning. However, its optimization problem is nonconvex and existing MMC methods all rely on reformulating and relaxing the nonconvex optimization problem as semidefinite programs (SDP). Though SDP is convex and standard solvers are available, they are computationally very expensive and only small data sets can be handled. To make MMC more practical, we avoid SDP relaxations and propose in this paper an efficient approach that performs alternating optimization directly on the original nonconvex problem. A key step to avoid premature convergence in the resultant iterative procedure is to change the loss function from the hinge loss to the Laplacian/square loss so that overconfident predictions are penalized. Experiments on a number of synthetic and real-world data sets demonstrate that the proposed approach is more accurate, much faster (hundreds to tens of thousands of times faster), and can handle data sets that are hundreds of times larger than the largest data set reported in the MMC literature.     

历史信息学习的相关反馈图像检索

相关反馈技术被有效的应用于基于内容的图像检索.传统的相关反馈未能充分利用检索的历史信息.为了进一步提高检索的效率与准确性,提出一种基于历史检索信息学习的相关反馈检索方法.该方法将每次检索的结果作为历史检索信息保存.进行新的检索时,判断当前查询图像与历史检索信息的语义相关性,预测检索结果,以期减少相关反馈次数.对包含80 00幅图像的图像库实验表明,与传统相关反馈技术相比,该方法明显的改善了检索性能.