Unsupervised Learning of Foreground Object Segmentation

International Journal of Computer Vision - Unsupervised learning represents one of the most interesting challenges in computer vision today. The task has an immense practical value with many...     

融合无监督特征的藏文分词方法研究

藏文分词是藏文信息处理的基础性关键问题,目前基于序列标注的藏文分词方法大都采用音节位置特征和类别特征等。该文从无标注语料中抽取边界熵特征、邻接变化数特征、无监督间隔标注等无监督特征,并将之融合到基于序列标注的分词系统中。从实验结果可以看出,与基线藏文分词系统相比,分词F值提高了0.97%,并且未登录词识别结果也有较大的提高。说明,该文从无标注数据中提取出的无监督特征较为有效,和有监督的分词模型融合到一起显著提高了基线分词系统的效果。     

基于运动引导的高效无监督视频目标分割网络

大量基于深度学习的无监督视频目标分割(Unsupervised video object segmentation, UVOS)算法存在模型参数量与计算量较大的问题, 这显著限制了算法在实际中的应用. 提出了基于运动引导的视频目标分割网络, 在大幅降低模型参数量与计算量的同时, 提升视频目标分割性能. 整个模型由双流网络、运动引导模块、多尺度渐进融合模块三部分组成. 具体地, 首先, RGB图像与光流估计输入双流网络提取物体外观特征与运动特征; 然后, 运动引导模块通过局部注意力提取运动特征中的语义信息, 用于引导外观特征学习丰富的语义信息; 最后, 多尺度渐进融合模块获取双流网络的各个阶段输出的特征, 将深层特征渐进地融入浅层特征, 最终提升边缘分割效果. 在3个标准数据集上进行了大量评测, 实验结果表明了该方法的优越性能.     

Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation

We present a method to learn probabilistic object models (POMs) with minimal supervision, which exploit different visual cues and perform tasks such as classification, segmentation, and recognition. We formulate this as a structure induction and learning task and our strategy is to learn and combine elementary POMs that make use of complementary image cues. We describe a novel structure induction procedure, which uses knowledge propagation to enable POMs to provide information to other POMs and “teach them” (which greatly reduces the amount of supervision required for training and speeds up the inference). In particular, we learn a POM-IP defined on Interest Points using weak supervision [1], [2] and use this to train a POM-mask, defined on regional features, which yields a combined POM that performs segmentation/localization. This combined model can be used to train POM-edgelets, defined on edgelets, which gives a full POM with improved performance on classification. We give detailed experimental analysis on large data sets for classification and segmentation with comparison to other methods. Inference takes five seconds while learning takes approximately four hours. In addition, we show that the full POM is invariant to scale and rotation of the object (for learning and inference) and can learn hybrid objects classes (i.e., when there are several objects and the identity of the object in each image is unknown). Finally, we show that POMs can be used to match between different objects of the same category, and hence, enable objects recognition.     

维吾尔文无监督自动切分及无监督特征选择

维吾尔文常用切分方法会产生大量的语义抽象甚至多义的词特征,因此学习算法难以发现高维数据中隐藏的结构.提出一种无监督切分方法dme-TS和一种无监督特征选择方法UMRMR-UFS.dme-TS从大规模生语料中自动获取单词Bi-gram及上下文语境信息,并将相邻单词间的t-测试差、互信息及双词上下文邻接对熵的线性融合作为一个组合统计量(dme)来评价单词间的结合能力,从而将文本切分成语义具体的独立语言单位的特征集合.UMRMR-UFS用一种综合考虑最大相关度和最小冗余的无监督特征选择标准(UMRMR)来评价每一个特征的重要性,并将最重要的特征依次移入到特征子集中.实验结果表明dme-TS能有效控制原始特征集的规模,提高特征项本身的质量,用UMRMR-UFS的输出来表征文本时,学习算法也表现出其最高的性能.     

Learning Recognition and Segmentation Using the Cresceptron

This paper presents a framework called Cresceptron for view-based learning, recognition and segmentation. Specifically, it recognizes and segments image patterns that are similar to those learned, using a stochastic distortion model and view-based interpolation, allowing other view points that are moderately different from those used in learning. The learning phase is interactive. The user trains the system using a collection of training images. For each training image, the user manually draws a polygon outlining the region of interest and types in the label of its class. Then, from the directional edges of each of the segmented regions, the Cresceptron uses a hierarchical self-organization scheme to grow a sparsely connected network automatically, adaptively and incrementally during the learning phase. At each level, the system detects new image structures that need to be learned and assigns a new neural plane for each new feature. The network grows by creating new nodes and connections which memorize the new image structures and their context as they are detected. Thus, the structure of the network is a function of the training exemplars. The Cresceptron incorporates both individual learning and class learning; with the former, each training example is treated as a different individual while with the latter, each example is a sample of a class. In the performance phase, segmentation and recognition are tightly coupled. No foreground extraction is necessary, which is achieved by backtracking the response of the network down the hierarchy to the image parts contributing to recognition. Several stochastic shape distortion models are analyzed to show why multilevel matching such as that in the Cresceptron can deal with more general stochastic distortions that a single-level matching scheme cannot. The system is demonstrated using images from broadcast television and other video segments to learn faces and other objects, and then later to locate and to recognize similar, but possibly distorted, views of the same objects.     

融合颜色与纹理的复杂场景下的服装图像分割算法

针对复杂场景下拍摄到的服装图像的分割问题,提出一种基于先验知识的融合颜色和纹理特征的无监督分割算法。首先利用块截断编码思想将传统的三维颜色空间截断成为六维空间,得到更为精细的颜色特征,并结合改进的局部二值模式纹理特征实现对图像的特征描述;然后根据目标区域和背景区域在图像中出现的统计规律,提出了一种基于先验知识的两分法来对图像进行分割。由于对图像做了分块处理,因此在子图像块的基础上进行的图像分割将更加高效。实验表明,设计的算法能快速有效地将目标区域从各类不同的复杂场景中分割出来,且整个过程无须人工设定任何参数,对后续的图像理解和图像检索具有重要意义。     

Learning Kernels for Unsupervised Domain Adaptation with Applications to Visual Object Recognition

Domain adaptation aims to correct the mismatch in statistical properties between the source domain on which a classifier is trained and the target domain to which the classifier is to be applied. In this paper, we address the challenging scenario of unsupervised domain adaptation, where the target domain does not provide any annotated data to assist in adapting the classifier. Our strategy is to learn robust features which are resilient to the mismatch across domains and then use them to construct classifiers that will perform well on the target domain. To this end, we propose novel kernel learning approaches to infer such features for adaptation. Concretely, we explore two closely related directions. In the first direction, we propose unsupervised learning of a geodesic flow kernel (GFK). The GFK summarizes the inner products in an infinite sequence of feature subspaces that smoothly interpolates between the source and target domains. In the second direction, we propose supervised learning of a kernel that discriminatively combines multiple base GFKs. Those base kernels model the source and the target domains at fine-grained granularities. In particular, each base kernel pivots on a different set of landmarks—the most useful data instances that reveal the similarity between the source and the target domains, thus bridging them to achieve adaptation. Our approaches are computationally convenient, automatically infer important hyper-parameters, and are capable of learning features and classifiers discriminatively without demanding labeled data from the target domain. In extensive empirical studies on standard benchmark recognition datasets, our appraches yield state-of-the-art results compared to a variety of competing methods.     

基于无监督提取表情时空特征的情感识别

情感识别是解决智能教学系统中情感缺失问题的关键技术。针对识别时如何从视频中有效提取人脸表情时空特征的问题,提出一种采用堆叠卷积独立子空间分析模型进行无监督特征提取的识别方法,来对疑惑、愉快和厌倦3种学习中最常出现的情感进行识别。该方法检测视频中的人脸区域并进行规范化处理,采用堆叠卷积独立子空间分析模型从视频块中无监督地学习表情的时空特征,采用线性支持向量机进行分类。实验结果表明,相比使用人工特征的方法,该方法能够更有效地提取视频中人脸表情的时空特征,获得更高的识别率,同时符合实时性要求。     

基于无指导学习策略的无词表条件下的汉语自动分词

探讨了基于无指导学习策略和无词表条件下的汉语自动分词方法,以期对研制开放环境下健壮的分词系统有所裨益,全部分词知识源自从生语料库中自动获得的汉字Bigram．在字间互信息和t-测试差的基础上,提出了一种将两者线性叠加的新的统计量md,并引入了峰和谷的概念,进而设计了相应的分词算法,大规模开放测试结果显示,该算法关于字间位置的分词正确率为85．88％,较单独使用互信息或t-测试差分别提高了2．47％和5．66％。     

Unsupervised Learning of Word Segmentation Rules with Genetic Algorithms and Inductive Logic Programming

This article presents a combination of unsupervised and supervised learning techniques for the generation of word segmentation rules from a raw list of words. First, a language bias for word segmentation is introduced and a simple genetic algorithm is used in the search for a segmentation that corresponds to the best bias value. In the second phase, the words segmented by the genetic algorithm are used as an input for the first order decision list learner CLOG. The result is a set of first order rules which can be used for segmentation of unseen words. When applied on either the training data or unseen data, these rules produce segmentations which are linguistically meaningful, and to a large degree conforming to the annotation provided.     

Statistical Learning by Imitation of Competing Constraints in Joint Space and Task Space

We present a probabilistic architecture for solving generically the problem of extracting the task constraints through a programming by demonstration framework and for generalizing the acquired knowledge to various situations. In previous work, we proposed an approach based on Gaussian mixture regression to find a controller for the robot reproducing the statistical characteristics of a movement in joint space and in task space through Lagrange optimization. In this paper, we develop an alternative procedure to handle simultaneously constraints in joint space and in task space by combining directly the probabilistic representation of the task constraints with a solution to Jacobian-based inverse kinematics. The method is validated in manipulation tasks with two 5-d.o.f. Katana robotic arms displacing a set of objects.     

Multiscale Annealing for Grouping and Unsupervised Texture Segmentation

We derive real-time global optimization methods for several clustering optimization problems commonly used in unsupervised texture segmentation. Speed is achieved by exploiting the image neighborhood relation of features to design a multiscale optimization technique, while accuracy and global optimization properties are gained using annealing techniques. Coarse grained cost functions are derived for central and histogram-based clustering as well as several sparse proximity-based clustering methods. For optimization deterministic annealing algorithms are applied. Annealing schedule, coarse-to-fine optimization and the estimated number of segments are tightly coupled by a statistical convergence criterion derived from computational learning theory. The notion of optimization scale parametrized by a computational temperature is thus unified with the scales defined by the image resolution and the model or segment complexity. The algorithms are benchmarked on Brodatz-like microtexture mixtures. Results are presented for an autonomous robotics application. Extensions are discussed in the context of prestructuring large image databases valuable for fast and reliable image retrieval.     

Information-Theoretic Active Polygons for Unsupervised Texture Segmentation

Curve evolution models used in image segmentation and based on image region information usually utilize simple statistics such as means and variances, hence can not account for higher order nature of the textural characteristics of image regions. In addition, the object delineation by active contour methods, results in a contour representation which still requires a substantial amount of data to be stored for subsequent multimedia applications such as visual information retrieval from databases. Polygonal approximations of the extracted continuous curves are required to reduce the amount of data since polygons are powerful approximators of shapes for use in later recognition stages such as shape matching and coding. The key contribution of this paper is the development of a new active contour model which nicely ties the desirable polygonal representation of an object directly to the image segmentation process. This model can robustly capture texture boundaries by way of higher-order statistics of the data and using an information-theoretic measure and with its nature of the ordinary differential equations. This new variational texture segmentation model, is unsupervised since no prior knowledge on the textural properties of image regions is used. Another contribution in this sequel is a new polygon regularizer algorithm which uses electrostatics principles. This is a global regularizer and is more consistent than a local polygon regularization in preserving local features such as corners.Supported by NSF grant CCR-0133736.Partially supported by AFOSR grant F49620-98-1-0190 and NSF grant CCR-9984067.     

非监督、多级嘴唇区域分割方法

该文介绍了一个非监督、多级嘴唇区域分割检测方法。首先提出利用fisher变换增强嘴唇区域,然后利用嘴唇在人脸区域的分布面积比,提出利用统计阈值完成嘴唇的初步分割,最后利用嘴唇对称性和轮廓光滑性的先验知识,提出了基于局部阈值调整完成嘴唇的精细分割。实验证明该方法在不同光照条件下、对不同人、各种表情都能自动地、鲁棒地、精确地检测出嘴唇区域,并较准确提取初步的嘴唇几何参数。利用这些几何参数作为轮廓定位的初始化条件,可以大大提高变形模板和ASM模型等嘴唇轮廓定位算法的速度和准确度。     

基于多模特征深度学习的机器人抓取判别方法

针对智能机器人抓取判别问题,研究多模特征深度学习与融合方法.该方法将测试特征分布偏离训练特征视为一类噪化,引入带稀疏约束的降噪自动编码（Denoising auto-encoding, DAE）,实现网络权值学习;并以叠层融合策略,获取初始多模特征的深层抽象表达,两种手段相结合旨在提高深度网络的鲁棒性和抓取判别精确性.实验采用深度摄像机与6自由度工业机器人组建测试平台,对不同类别目标进行在线对比实验.结果表明,设计的多模特征深度学习依据人的抓取习惯,实现最优抓取判别,并且机器人成功实施抓取定位,研究方法对新目标具备良好的抓取判别能力.     

Unsupervised Learning for Graph Matching

Graph matching is an essential problem in computer vision that has been successfully applied to 2D and 3D feature matching and object recognition. Despite its importance, little has been published on learning the parameters that control graph matching, even though learning has been shown to be vital for improving the matching rate. In this paper we show how to perform parameter learning in an unsupervised fashion, that is when no correct correspondences between graphs are given during training. Our experiments reveal that unsupervised learning compares favorably to the supervised case, both in terms of efficiency and quality, while avoiding the tedious manual labeling of ground truth correspondences. We verify experimentally that our learning method can improve the performance of several state-of-the art graph matching algorithms. We also show that a similar method can be successfully applied to parameter learning for graphical models and demonstrate its effectiveness empirically.     

行人再识别中基于无监督学习的粗细粒度特征提取

行人再识别研究中存在特征判别信息不够丰富的情况,并且遮挡、光照等因素会干扰有效特征的准确提取,对后续相似性度量、度量结果排序等工作都有较大影响。此外,监督学习需要使用标签信息,在面对大型数据集时工作量很大。通过引入无监督学习框架,提出一种粗细粒度判别性特征提取方法。构建基于细粒度和粗粒度特征学习的模型框架,其中包含局部和全局2个分支。在局部分支中,对图像学习到的特征映射提取补丁块,并在未标记数据集上学习不同位置的细粒度补丁特征;在全局分支中,使用无标注数据集的相似度和多样性作为信息来学习粗粒度特征。在此基础上,利用相吸和相斥2个损失函数分别增加类别内相似度和类别间多样性,并结合最小距离准则计算特征之间的相似度,进行无监督的聚类合并。在Market-1501和DukeMTMC-reID数据集上的实验结果表明,该方法对于完成行人再识别任务具有较好的判别性能和鲁棒性,相比所有对比方法的最优结果,其Rank-1指标分别提高5.76%和5.07%,平均精度均值分别提高3.2%和5.6%。     

一种两层次无监督的音频分割算法

本文提出一种两层次无监督音频分割算法,它用于检测音频流中的说话人、环境、信道等声学特征变化点,该方法将音频分割过程分为两个层次: 区域层次和边界层次,通过固定检测窗移动,它快速定位到声学特征变化点存在的区域,然后在潜在变化区域内采用T² 统计值和贝叶斯信息准则(BIC)结合的方法快速确定片断边界。在区域检测层次,将修正的广义对数似然比准则应用于潜在的变化区域检测,它即无需设定阈值门限又可保证低的漏检率,在1997年Hub4中文广播语音数据库上的实验结果表明,同传统的混合分割算法比较,该算法在处理速度得到提高的同时,声学特征变化点的召回率提高10.5％。