Space-Time Tree Ensemble for Action Recognition and Localization

Human actions are, inherently, structured patterns of body movements. We explore ensembles of hierarchical spatio-temporal trees, discovered directly from training data, to model these structures for action recognition and spatial localization. Discovery of frequent and discriminative tree structures is challenging due to the exponential search space, particularly if one allows partial matching. We address this by first building a concise action word vocabulary via discriminative clustering of the hierarchical space-time segments, which is a two-level video representation that captures both static and non-static relevant space-time segments of the video. Using this vocabulary we then utilize tree mining with subsequent tree clustering and ranking to select a compact set of discriminative tree patterns. Our experiments show that these tree patterns, alone, or in combination with shorter patterns (action words and pairwise patterns) achieve promising performance on three challenging datasets: UCF Sports, HighFive and Hollywood3D. Moreover, we perform cross-dataset validation, using trees learned on HighFive to recognize the same actions in Hollywood3D, and using trees learned on UCF-Sports to recognize and localize the similar actions in JHMDB. The results demonstrate the potential for cross-dataset generalization of the trees our approach discovers.     

Statistical Modeling of the 3D Geometry and Topology of Botanical Trees

We propose a framework for statistical modeling of the 3D geometry and topology of botanical trees. We treat botanical trees as points in a tree‐shape space equipped with a proper metric that captures the geometric and the topological differences between trees. Geodesics in the tree‐shape space correspond to the optimal sequence of deformations, i.e. bending, stretching, and topological changes, which align one tree onto another. In this way, the 3D tree modeling and synthesis problem becomes a problem of exploring the tree‐shape space either in a controlled fashion, using statistical regression, or randomly by sampling from probability distributions fitted to populations in the tree‐shape space. We show how to use this framework for (1) computing statistical summaries, e.g. the mean and modes of variations, of a population of botanical trees, (2) synthesizing random instances of botanical trees from probability distributions fitted to a population of botanical trees, and (3) modeling, interactively, 3D botanical trees using a simple sketching interface. The approach is fast and only requires as input 3D botanical tree models with a known upright orientation.     

基于自适应布告板的三维树木表达方法

植被、地形以及人工建构筑物是三维虚拟地理场景可视化表达的基本内容,树木是植被的主要组成部分.由于树木自然形态的复杂性,其真实感表达非常困难.基于几何模型的树木表达可以产生逼真的细节,但场景实时渲染具有巨大的计算负担,而基于纹理的简化模型在真实感表达上有所欠缺,但具有更好的渲染性能.如何兼顾场景渲染效率与视觉真实感受一直是树木三维可视化表达的研究热点.本文提出一种基于自适应布告板的三维树木表达方法,该方法在通常的平面布告板方法基础上,根据视点与布告板的相对位置关系,从预先获得的多张树木影像中选取最符合视点与树木相对位置关系的影像作为树木纹理渲染于该布告板上,实现了布告板纹理的动态调整,使得不同视点下的树木渲染效果尽量接近真实,同时也具有布告板的渲染性能优势.本文基于WebGL在浏览器中构建了一个三维场景,其中包含若干树木模型.实验结果表明本方法在表达树林时,在浏览器这样的低渲染计算能力环境下也能取得可接受的渲染性能与表达效果.     

Texture mixing and texture movie synthesis using statisticallearning

We present an algorithm based on statistical learning for synthesizing static and time-varying textures matching the appearance of an input texture. Our algorithm is general and automatic and it works well on various types of textures, including 1D sound textures, 2D texture images, and 3D texture movies. The same method is also used to generate 2D texture mixtures that simultaneously capture the appearance of a number of different input textures. In our approach, input textures are treated as sample signals generated by a stochastic process. We first construct a tree representing a hierarchical multiscale transform of the signal using wavelets. From this tree, new random trees are generated by learning and sampling the conditional probabilities of the paths in the original tree. Transformation of these random trees back into signals results in new random textures. In the case of 2D texture synthesis, our algorithm produces results that are generally as good as or better than those produced by previously described methods in this field. For texture mixtures, our results are better and more general than those produced by earlier methods. For texture movies, we present the first algorithm that is able to automatically generate movie clips of dynamic phenomena such as waterfalls, fire flames, a school of jellyfish, a crowd of people, etc. Our results indicate that the proposed technique is effective and robust     

Inverse Procedural Modelling of Trees

Procedural tree models have been popular in computer graphics for their ability to generate a variety of output trees from a set of input parameters and to simulate plant interaction with the environment for a realistic placement of trees in virtual scenes. However, defining such models and their parameters is a difficult task. We propose an inverse modelling approach for stochastic trees that takes polygonal tree models as input and estimates the parameters of a procedural model so that it produces trees similar to the input. Our framework is based on a novel parametric model for tree generation and uses Monte Carlo Markov Chains to find the optimal set of parameters. We demonstrate our approach on a variety of input models obtained from different sources, such as interactive modelling systems, reconstructed scans of real trees and developmental models.     

基于Lobe表示的树木快速3D建模

提出了一种基于Lobe表示的树木三维建模方法,该方法的核心思想在于树冠形状取决于外形轮廓,而其内部的枝叶结构具有相似性,可以从有限的模板库通过纹理合成的方法近似生成.该方法可以描述成两大部分:1)编码.将给定树木表示成骨架和Lobe的形式;2)解码.根据Lobe形状,基于已有的枝叶模板生成Lobe内部三维结构.这种基于Lobe的轻量级树木表示方法在很大程度上提高了大规模树木三维模型存储和传输的效率,非常适用于网络环境下大规模城市场景的交互式应用.     

风场作用下的动态森林场景的实时仿真

为真实地模拟树在风中摇曳的动态效果,将风场视为一个随机过程(场),利用谱分析的方法生成给定条件下的风速矢量场;进而采用简化的动力学方程,实时计算风力作用下树的位移和形变.采用一种混合式几何和图像的表示方法对树木进行参数化建模,并在模型中内建多个细节层次,以便加速场景视点相关的绘制.综合以上算法的交互式漫游系统能实时地模拟包含数千棵树的森林场景在风场作用下的动态效果,初步满足了虚拟现实、战场仿真与数字娱乐等应用的要求.     

A Steady State Analysis of Diffracting Trees

Diffracting trees are an effective and highly scalable distributed- parallel technique for shared counting and load balancing. This paper presents the first steady-state combinatorial model and analysis for diffracting trees, and uses it to answer several critical algorithmic design questions. Our model is simple and sufficiently high level to overcome many implementation-specific details, and yet as we show it is rich enough to predict empirically observed behaviors accurately. As a result of our analysis we were able to identify starvation problems in the original diffracting tree algorithm and modify it to a create a more stable version. We were also able to identify the range in which the diffracting tree performs most efficiently, and the ranges in which its performance degrades. We believe our model and modeling approach open the way to steady-state analysis of other distributed-parallel structures such as counting networks and elimination trees. Received October 1996, and in final form July 1997.     

统计句法分析建模中基于信息论的特征类型分析

统计句法分析利用概率评价模型评价每棵选句法树存在的可能性,选择概率值最高的候选句法树作为最终的句法分析结果。因此,统计句法分析的核心是一个概率评价模型,而各种概率评价模型的本质区别主要在于它们分别是根据上下文中的哪些特征来赋予句法树概率的。在统计句法分析研究领域,虽然已经提出了大量的概率评价模型,然而,不同的模型用得到了不同类型的特征,如何评价这些特征类型对于句法分析的作用呢？针对以上的问题,本研究为统计句法分析提出了一种特征类型的分析模型,该模型可以从信息论的角度量化地分析不同类型的上下文特征对于句法结构的预测作用。其基本思想是利用信息论中熵与条件熵的度量来显示一个特征类型是否抓住了预测句法结构的主要信息。如果加入某个特征类型之后当前句法结构的不确定性（熵）明显下降,则认为该特征类型抓住了上下文中影响句法结构的某些主要信息。特征类型分析的信息论模型利用预测信息量、预测信息增益、预测信息关联度以及预测信息总量四种度量从不同的仙量化地分析各种特征类型及特征类型组合对于当前目标的预测作用。实验以Penn TreeBank为训练集,将上下文中不同的特征类型对于句法分析规则的预测作用进行了系统的量化分析,得出了一系列有关不同特征类型及特征类型组合对句法结构的预测作用的结论。     

Image-based tree modeling from a few images with very narrow viewing range

Creating 3D tree models from actual trees is a task receiving increasing attention. Some approaches have been developed to reconstruct a tree based on a number of photographs around the tree, typically spanning a wide viewing range. However, due to the environmental restrictions, sometimes it is quite difficult to capture so many acceptable images from so many different viewpoints. In this paper, we propose a tree modeling system which is capable of reconstructing the 3D model of a tree from a few images with very narrow viewing ranges. Because only a few images are required to generate the model, our system has the distinct advantage of fewer environmental restrictions, resulting in the extended usability and flexibility for real applications.

Inference of tree automata from sample set of trees

A method for inferring of tree automata from sample set of trees is presented. The procedure, which is based on the concept ofk-follower of a tree with respect to the sample tree set, produces a tree automaton capable of accepting all the sample trees as well as other trees similar in structure. The behavior of the inferred tree automaton for varying values of parameterk is also discussed.This work was supported in part by a Scientific Research Grant-In-Aid (Grant No. 57460129) from the Ministry of Education, Science and Culture, Japan.     

基于机载激光雷达点云的交互式树木分割与建模方法研究

机载激光雷达扫描(ALS)系统可大规模获取地表树木点云,有助于较高精度树木结构参数提取 和景观层面的几何重建,然而树木复杂的拓扑结构和树种的多样性给树木的准确分割与建模带来挑战。传统基 于点云的自动树木分割和建模算法虽然效率高,但存在分割误差较大、建模鲁棒性较差等问题,难以满足深度 学习大背景下用户对树木分割与建模结果进行精准标注的需求。针对 ALS 树点云密度低、点云缺失导致的自 动分割与建模困难等问题,提出一种基于高度图的交互式层次分割法用于从稀疏树点云中提取单棵树点云,然 后基于改进的空间殖民算法,通过交互式调节约束角度、删除阈值和影响半径等 3 个参数实现单棵树的重建。 实验结果表明,提出的交互式分割算法能够解决最小生成树和规范割等分割算法产生的误分割问题;提出的基 于改进空间殖民算法的交互式重建算法鲁棒性好,能够有效解决稀疏及缺失树点云的几何重建问题,且生成的 树模型能够较好地保留原始点云的特征。     

Improved Binary Space Partition merging

This paper presents a new method for evaluating boolean set operations between Binary Space Partition (BSP) trees. Our algorithm has many desirable features, including both numerical robustness and O(n) output sensitive time complexity, while simultaneously admitting a straightforward implementation. To achieve these properties, we present two key algorithmic improvements. The first is a method for eliminating null regions within a BSP tree using linear programming. This replaces previous techniques based on polygon cutting and tree splitting. The second is an improved method for compressing BSP trees based on a similar approach within binary decision diagrams. The performance of the new method is analyzed both theoretically and experimentally. Given the importance of boolean set operations, our algorithms can be directly applied to many problems in graphics, CAD and computational geometry.     

3D树木建模技术研究进展

3D树木建模一直都是计算机图形学、计算机视觉、虚拟现实等领域最具有挑战性的研究方向之一,国外学者在该领域做了大量研究工作,但3D树木建模综述性文章的缺乏成为其发展的制约因素。侧重从计算机视觉的角度,将当前国内外经典的3D树木建模方法分为基于图像,基于规则和基于草图等三类,并跟踪了最新的3D树木建模技术进展,对他们的成果进行了深入剖析,重点阐述3D树木建模关键技术。最后,分析和比较了几种主要的3D树木建模技术,提出轻量化3D树木建模技术是今后虚拟场景中的交互应用的发展趋势。此外,根据当前研究的难点提出了基于单张图像的轻量化混合建模的研究思路。     

Similarity K-d tree method for sparse point pattern matching with underlying non-rigidity

We propose a method for matching non-affinely related sparse model and data point-sets of identical cardinality, similar spatial distribution and orientation. To establish a one-to-one match, we introduce a new similarity K-dimensional tree. We construct the tree for the model set using spatial sparsity priority order. A corresponding tree for the data set is then constructed, following the sparsity information embedded in the model tree. A matching sequence between the two point sets is generated by traversing the identically structured trees. Experiments on synthetic and real data confirm that this method is applicable to robust spatial matching of sparse point-sets under moderate non-rigid distortion and arbitrary scaling, thus contributing to non-rigid point-pattern matching.     

基于交互变形的树木三维建模研究

通过对树枝的交互变形处理,提供了一种能美化树木形状的三维建模方法。该方法由三部分组成：基于交互变形的树木三维表示模型、树枝的交互选择机制及对所选择树枝的变形处理。树枝变形分基本为形和综合变形两种,基本变表提供对树枝增删、伸缩、弯曲、旋转等简单变表操作;综合变形以基本变形为基础、以美学理论为准则对对木形状进行整体的美学控制。文中提供的方法既可与现有树木生长建模方法结合使用,也可独立运用,以达到对树木形状的建模和美化处理。     

A data mining proxy approach for efficient frequent itemset mining

Data mining has attracted a lot of research efforts during the past decade. However, little work has been reported on the efficiency of supporting a large number of users who issue different data mining queries periodically when there are new needs and when data is updated. Our work is motivated by the fact that the pattern-growth method is one of the most efficient methods for frequent pattern mining which constructs an initial tree and mines frequent patterns on top of the tree. In this paper, we present a data mining proxy approach that can reduce the I/O costs to construct an initial tree by utilizing the trees that have already been resident in memory. The tree we construct is the smallest for a given data mining query. In addition, our proxy approach can also reduce CPU cost in mining patterns, because the cost of mining relies on the sizes of trees. The focus of the work is to construct an initial tree efficiently. We propose three tree operations to construct a tree. With a unique coding scheme, we can efficiently project subtrees from on-disk trees or in-memory trees. Our performance study indicated that the data mining proxy significantly reduces the I/O cost to construct trees and CPU cost to mine patterns over the trees constructed.     

Combining shape and physical modelsfor online cursive handwriting synthesis

This paper proposes a novel learning-based approach to synthesizing cursive handwriting of a user's personal handwriting style by combining shape and physical models. In the training process, some sample paragraphs written by a user are collected and these cursive handwriting samples are segmented into individual characters by using a two-level writer-independent segmentation algorithm. Samples for each letter are then aligned and trained using shape models. In the synthesis process, a delta log-normal model based conditional sampling algorithm is proposed to produce smooth and natural cursive handwriting of the user's style from models. Received: 26 April 2003, Accepted: 27 September 2004, Published online: 29 November 2004 Correspondence to: Jue Wang Jue Wang and Chenyu Wu completed this work while interns at Microsoft Research Asia.     

Embedding ofk-ary Complete Trees into Hypercubes with Uniform Load

The main result of this paper is an algorithm for embeddingk-ary complete trees into hypercubes with uniform load and asymptotically optimal dilation. The algorithm is fully scalable—the dimension of the hypercube can be chosen independent of the arity and height of the tree. The embedding enables to emulate optimally both Divide&Conquer computations onk-ary complete trees where only one level of nodes is active at a time and general computations based onk-ary complete trees where all tree nodes are active simultaneously. As a special case, we get an algorithm for embedding thek-ary complete tree of given height into its optimal hypercube with load 1 and asymptotically optimal dilation.     

Classification tree analysis using TARGET

Tree models are valuable tools for predictive modeling and data mining. Traditional tree-growing methodologies such as CART are known to suffer from problems including greediness, instability, and bias in split rule selection. Alternative tree methods, including Bayesian CART (Chipman et al., 1998; Denison et al., 1998), random forests (Breiman, 2001a), bootstrap bumping (Tibshirani and Knight, 1999), QUEST (Loh and Shih, 1997), and CRUISE (Kim and Loh, 2001), have been proposed to resolve these issues from various aspects, but each has its own drawbacks.Gray and Fan (2003) described a genetic algorithm approach to constructing decision trees called tree analysis with randomly generated and evolved trees (TARGET) that performs a better search of the tree model space and largely resolves the problems with current tree modeling techniques. Utilizing the Bayesian information criterion (BIC), Fan and Gray (2005) developed a version of TARGET for regression tree analysis. In this article, we consider the construction of classification trees using TARGET. We modify the BIC to handle a categorical response variable, but we also adjust its penalty component to better account for the model complexity of TARGET. We also incorporate the option of splitting rules based on linear combinations of two or three variables in TARGET, which greatly improves the prediction accuracy of TARGET trees. Comparisons of TARGET to existing methods, using simulated and real data sets, indicate that TARGET has advantages over these other approaches.