Semi-supervised video segmentation using tree structured graphical models

We present a novel, implementation friendly and occlusion aware semi-supervised video segmentation algorithm using tree structured graphical models, which delivers pixel labels alongwith their uncertainty estimates. Our motivation to employ supervision is to tackle a task-specific segmentation problem where the semantic objects are pre-defined by the user. The video model we propose for this problem is based on a tree structured approximation of a patch based undirected mixture model, which includes a novel time-series and a soft label Random Forest classifier participating in a feedback mechanism. We demonstrate the efficacy of our model in cutting out foreground objects and multi-class segmentation problems in lengthy and complex road scene sequences. Our results have wide applicability, including harvesting labelled video data for training discriminative models, shape/pose/articulation learning and large scale statistical analysis to develop priors for video segmentation. © 2011 IEEE.

Diffusive coevolution and mutualism maintenance mechanisms in a fig-fig wasp system

In reciprocal mutualism systems, the exploitation events by exploiters might disrupt the reciprocal mutualism, wherein one exploiter species might even exclude other coexisting exploiter species over an evolutionary time frame. What remains unclear is how such a community is maintained. Niche partitioning, or spatial heterogeneity among the mutualists and exploiters, is generally believed to enable stability within a mutualistic system. However, our examination of a reciprocal mutualism between a fig species (Ficus racemosa) and its pollinator wasp (Ceratosolen fusciceps) shows that spatial niche partitioning does not sufficiently prevent exploiters from overexploiting the common resource (i.e., the female flowers), because of the considerable niche overlap between the mutualists and exploiters. In response to an exploiter, our experiment shows that the fig can (1) abort syconia-containing flowers that have been galled by the exploiter, Apocryptophagus testacea, which oviposits before the pollinators do; and (2) retain syconia-containing flowers galled by Apocryptophagus mayri, which oviposit later than pollinators. However, as a result of (2), there is decreased development of adult non-pollinators or pollinator species in syconia that have not been sufficiently pollinated, but not aborted. Such discriminative abortion of figs or reduction in offspring development of exploiters while rewarding cooperative individuals with higher offspring development by the fig will increase the fitness of cooperative pollinating wasps, but decrease the fitness of exploiters. The fig fig wasp interactions are diffusively coevolved, a case in which fig wasps diversify their genotype, phenotype, or behavior as a result of competition between wasps, while figs diverge their strategies to facilitate the evolution of cooperative fig waps or lessen the detrimental behavior by associated fig wasps. In habitats or syconia that suffer overexploitation, discriminative abortion of figs or reduction in the offspring development of exploiters in syconia that are not or not sufficiently pollinated will decrease exploiter fitness and perhaps even drive the population of exploiters to local extinction, enabling the evolution and maintenance of cooperative pollinators through the movement between habitats or syconia (i.e., the metapopulations).

The evolution of cooperation in asymmetric systems

Explaining "Tragedy of the Commons" of evolution of cooperation remains one of the greatest problems for both biology and social science. Asymmetrical interaction, which is one of the most important characteristics of cooperative system, has not been sufficiently considered in the existing models of the evolution of cooperation. Considering the inequality in the number and payoff between the cooperative actors and recipients in cooperation systems, discriminative density-dependent interference competition will occur in limited dispersal systems. Our model and simulation show that the local but not the global stability of a cooperative interaction can be maintained if the utilization of common resource remains unsaturated, which can be achieved by density-dependent restraint or competition among the cooperative actors. More intense density dependent interference competition among the cooperative actors and the ready availability of the common resource, with a higher intrinsic contribution ratio of a cooperative actor to the recipient, will increase the probability of cooperation. The cooperation between the recipient and the cooperative actors can be transformed into conflict and, it oscillates chaotically with variations of the affecting factors under different environmental or ecological conditions. The higher initial relatedness (i.e. similar to kin or reciprocity relatedness), which is equivalent to intrinsic contribution ratio of a cooperative actor to the recipient, can be selected for by penalizing less cooperative or cheating actors but rewarding cooperative individuals in asymmetric systems. The initial relatedness is a pivot but not the aim of evolution of cooperation. This explains well the direct conflict observed in almost all cooperative systems.

Study morphologic, phylogenic relationships DNA barcoding of of identified decapod group (Crustacean) of Persian Gulf and Oman Sea.

Phylogenetic relationships among all described species (total of 12 taxa) of the decapoda, were examined with nucleotide sequence data from portions of mitochondrial gene and cytochrome oxidase subunit I (COI). The previous works on phylogeny proved that the mitochondrial COI gene in crustacean is a good discriminative marker at both inter- and intra-specific levels. We provide COI barcode sequences of commertial decapoda of Oman Sea, Persian Gulf, Iran. Industrial activities, ecologic considerations, and goals of the decapoda Barcode of Life campaign make it crucial that species of the south costal be identified. The reconstruction of evolut phylogeny of these species are crucial for revealing stock identity that can be used for the management of fisheries industries in Iran. Mitochondrial DNA sequences were used to reconstruct the phylogeny of the Penaeus species of marine shrimp. For this purpose, DNA was extracted using phenol- chloroform well as CTAB method. The evolutionary relationships among 12 species of the decapoda were examined using 610 bp of mitochondrial (mt) DNA from the cytochrome oxidase subunit I gene. Finally the cladograms were compared and the resulted phylogenetic trees confirmed that the Iran's species origin is Indo-west pacific species. Iran's species, which were not grouped with the other decapoda taxa seem to always form a sister clade with Indo-west pacific species with strong bootstrap support 100%. The result completely agrees with the previously defined species using morphological characters.However, we still lack any comprehensive and clear understanding of phylogenetic relationships in this group.

Neuronal activity in dorsomedial frontal cortex and prefrontal cortex reflecting irrelevant stimulus dimensions

Previous studies of the dorsomedial frontal cortex (DMF) and the prefrontal cortex (PF) have shown that, when monkeys respond to nonspatial features of a discriminative stimulus (e.g., color) and the stimulus appears at a place unrelated to the movement t

Derivative Kernels for Noise Robust ASR

Recently there has been interest in combined gen- erative/discriminative classifiers. In these classifiers features for the discriminative models are derived from generative kernels. One advantage of using generative kernels is that systematic approaches exist how to introduce complex dependencies beyond conditional independence assumptions. Furthermore, by using generative kernels model-based compensation/adaptation tech- niques can be applied to make discriminative models robust to noise/speaker conditions. This paper extends previous work with combined generative/discriminative classifiers in several directions. First, it introduces derivative kernels based on context- dependent generative models. Second, it describes how derivative kernels can be incorporated in continuous discriminative models. Third, it addresses the issues associated with large number of classes and parameters when context-dependent models and high- dimensional features of derivative kernels are used. The approach is evaluated on two noise-corrupted tasks: small vocabulary AURORA 2 and medium-to-large vocabulary AURORA 4 task.

Derivative kernels for noise robust ASR

Recently there has been interest in combining generative and discriminative classifiers. In these classifiers features for the discriminative models are derived from the generative kernels. One advantage of using generative kernels is that systematic approaches exist to introduce complex dependencies into the feature-space. Furthermore, as the features are based on generative models standard model-based compensation and adaptation techniques can be applied to make discriminative models robust to noise and speaker conditions. This paper extends previous work in this framework in several directions. First, it introduces derivative kernels based on context-dependent generative models. Second, it describes how derivative kernels can be incorporated in structured discriminative models. Third, it addresses the issues associated with large number of classes and parameters when context-dependent models and high-dimensional feature-spaces of derivative kernels are used. The approach is evaluated on two noise-corrupted tasks: small vocabulary AURORA 2 and medium-to-large vocabulary AURORA 4 task. © 2011 IEEE.

Extending noise robust structured support vector machines to larger vocabulary tasks

This paper describes a structured SVM framework suitable for noise-robust medium/large vocabulary speech recognition. Several theoretical and practical extensions to previous work on small vocabulary tasks are detailed. The joint feature space based on word models is extended to allow context-dependent triphone models to be used. By interpreting the structured SVM as a large margin log-linear model, illustrates that there is an implicit assumption that the prior of the discriminative parameter is a zero mean Gaussian. However, depending on the definition of likelihood feature space, a non-zero prior may be more appropriate. A general Gaussian prior is incorporated into the large margin training criterion in a form that allows the cutting plan algorithm to be directly applied. To further speed up the training process, 1-slack algorithm, caching competing hypothesis and parallelization strategies are also proposed. The performance of structured SVMs is evaluated on noise corrupted medium vocabulary speech recognition task: AURORA 4. © 2011 IEEE.

Silhouette-based object phenotype recognition using 3D shape priors

This paper tackles the novel challenging problem of 3D object phenotype recognition from a single 2D silhouette. To bridge the large pose (articulation or deformation) and camera viewpoint changes between the gallery images and query image, we propose a novel probabilistic inference algorithm based on 3D shape priors. Our approach combines both generative and discriminative learning. We use latent probabilistic generative models to capture 3D shape and pose variations from a set of 3D mesh models. Based on these 3D shape priors, we generate a large number of projections for different phenotype classes, poses, and camera viewpoints, and implement Random Forests to efficiently solve the shape and pose inference problems. By model selection in terms of the silhouette coherency between the query and the projections of 3D shapes synthesized using the galleries, we achieve the phenotype recognition result as well as a fast approximate 3D reconstruction of the query. To verify the efficacy of the proposed approach, we present new datasets which contain over 500 images of various human and shark phenotypes and motions. The experimental results clearly show the benefits of using the 3D priors in the proposed method over previous 2D-based methods. © 2011 IEEE.

MCBoost: Multiple classifier boosting for perceptual co-clustering of images and visual features

We present a new co-clustering problem of images and visual features. The problem involves a set of non-object images in addition to a set of object images and features to be co-clustered. Co-clustering is performed in a way that maximises discrimination of object images from non-object images, thus emphasizing discriminative features. This provides a way of obtaining perceptual joint-clusters of object images and features. We tackle the problem by simultaneously boosting multiple strong classifiers which compete for images by their expertise. Each boosting classifier is an aggregation of weak-learners, i.e. simple visual features. The obtained classifiers are useful for object detection tasks which exhibit multimodalities, e.g. multi-category and multi-view object detection tasks. Experiments on a set of pedestrian images and a face data set demonstrate that the method yields intuitive image clusters with associated features and is much superior to conventional boosting classifiers in object detection tasks.

Use of contexts in language model interpolation and adaptation

Language models (LMs) are often constructed by building multiple individual component models that are combined using context independent interpolation weights. By tuning these weights, using either perplexity or discriminative approaches, it is possible to adapt LMs to a particular task. This paper investigates the use of context dependent weighting in both interpolation and test-time adaptation of language models. Depending on the previous word contexts, a discrete history weighting function is used to adjust the contribution from each component model. As this dramatically increases the number of parameters to estimate, robust weight estimation schemes are required. Several approaches are described in this paper. The first approach is based on MAP estimation where interpolation weights of lower order contexts are used as smoothing priors. The second approach uses training data to ensure robust estimation of LM interpolation weights. This can also serve as a smoothing prior for MAP adaptation. A normalized perplexity metric is proposed to handle the bias of the standard perplexity criterion to corpus size. A range of schemes to combine weight information obtained from training data and test data hypotheses are also proposed to improve robustness during context dependent LM adaptation. In addition, a minimum Bayes' risk (MBR) based discriminative training scheme is also proposed. An efficient weighted finite state transducer (WFST) decoding algorithm for context dependent interpolation is also presented. The proposed technique was evaluated using a state-of-the-art Mandarin Chinese broadcast speech transcription task. Character error rate (CER) reductions up to 7.3 relative were obtained as well as consistent perplexity improvements. © 2012 Elsevier Ltd. All rights reserved.

A confidence-based approach for improving keyword hypothesis scores

The task in keyword spotting (KWS) is to hypothesise times at which any of a set of key terms occurs in audio. An important aspect of such systems are the scores assigned to these hypotheses, the accuracy of which have a significant impact on performance. Estimating these scores may be formulated as a confidence estimation problem, where a measure of confidence is assigned to each key term hypothesis. In this work, a set of discriminative features is defined, and combined using a conditional random field (CRF) model for improved confidence estimation. An extension to this model to directly address the problem of score normalisation across key terms is also introduced. The implicit score normalisation which results from applying this approach to separate systems in a hybrid configuration yields further benefits. Results are presented which show notable improvements in KWS performance using the techniques presented in this work. © 2013 IEEE.

Kernelized log linear models for continuous speech recognition

Large margin criteria and discriminative models are two effective improvements for HMM-based speech recognition. This paper proposed a large margin trained log linear model with kernels for CSR. To avoid explicitly computing in the high dimensional feature space and to achieve the nonlinear decision boundaries, a kernel based training and decoding framework is proposed in this work. To make the system robust to noise a kernel adaptation scheme is also presented. Previous work in this area is extended in two directions. First, most kernels for CSR focus on measuring the similarity between two observation sequences. The proposed joint kernels defined a similarity between two observation-label sequence pairs on the sentence level. Second, this paper addresses how to efficiently employ kernels in large margin training and decoding with lattices. To the best of our knowledge, this is the first attempt at using large margin kernel-based log linear models for CSR. The model is evaluated on a noise corrupted continuous digit task: AURORA 2.0. © 2013 IEEE.

Biologically Inspired Tensor Features

According to the research results reported in the past decades, it is well acknowledged that face recognition is not a trivial task. With the development of electronic devices, we are gradually revealing the secret of object recognition in the primate's visual cortex. Therefore, it is time to reconsider face recognition by using biologically inspired features. In this paper, we represent face images by utilizing the C1 units, which correspond to complex cells in the visual cortex, and pool over S1 units by using a maximum operation to reserve only the maximum response of each local area of S1 units. The new representation is termed C1 Face. Because C1 Face is naturally a third-order tensor (or a three dimensional array), we propose three-way discriminative locality alignment (TWDLA), an extension of the discriminative locality alignment, which is a top-level discriminate manifold learning-based subspace learning algorithm. TWDLA has the following advantages: (1) it takes third-order tensors as input directly so the structure information can be well preserved; (2) it models the local geometry over every modality of the input tensors so the spatial relations of input tensors within a class can be preserved; (3) it maximizes the margin between a tensor and tensors from other classes over each modality so it performs well for recognition tasks and (4) it has no under sampling problem. Extensive experiments on YALE and FERET datasets show (1) the proposed C1Face representation can better represent face images than raw pixels and (2) TWDLA can duly preserve both the local geometry and the discriminative information over every modality for recognition.

数据流上的频繁项集挖掘技术研究

数据流是近年出现的一个新的应用类型，具有连续、无限、高速等特点。典型的数据流包括：无线传感器网络应用环境中由传感器传回的各种监测数据、股票交易所的股票价格信息、网络监测系统与道路交通监测系统的监测数据、电信部门的通话记录数据，以及网站的日志信息等。数据流的出现对传统的数据管理和挖掘技术提出了巨大的挑战。传统的数据挖掘技术往往对静态数据集合做多遍扫描，其时间和空间复杂度均较高，难以直接应用到数据流环境中。本文对数据流上的频繁项集挖掘问题做了深入研究，主要研究内容和创新性成果概述如下： 本文首先对频繁项集挖掘问题做了一个全面的综述。综述部分先对静态数据集上的频繁项集挖掘的概念、分类、经典算法等相关研究做全面的介绍，然后分析了在数据流上进行频繁项集挖掘面临的问题和挑战、以及研究现状。 针对数据流上的频繁元素挖掘问题，本文提出了一个简单而高效的算法，挖掘数据流滑动窗口上的频繁元素。算法既可以定期返回满足ε-近似要求的频繁元素，也可以响应用户在任意时间提交的请求，返回满足误差要求的结果。 针对数据流上的频繁项集挖掘问题，本文提出了BFI-Stream算法，挖掘数据流滑动窗口上的所有频繁项集，实时返回精确结果。该算法使用前缀树数据结构，并且在创建和更新过程中裁剪了一部分非频繁节点，因此算法的空间和时间复杂度都较低。 接着，本文针对现有的在数据流上挖掘频繁项集的算法存在维护过多非频繁项集而导致使用空间过大的问题，提出了一种乐观裁剪方法，大大降低了算法的空间复杂度。文中先对实际数据集分析了项集的频率分布情况，提出了乐观裁剪方法，裁剪大部分非频繁项集；实验结果表明乐观裁剪方法不仅大大降低了内存使用量，还提高了算法的更新效率。 再次，本文针对用户指定最小支持度和允许误差的近似查询，提出了在数据流滑动窗口上挖掘频繁项集的近似算法AFI-Stream，返回满足误差要求的结果。AFI-Stream仅仅维护频繁项集，不维护非频繁项集，因此能大大降低算法使用的内存。 为了满足在数据流上挖掘频繁项集研究的需要，本文设计并开发了一个数据流频繁项集挖掘原型系统StreamMiner，进行相关算法的评测和研究。