Evidence-based recognition of 3-D objects

An evidence-based recognition technique is defined that identifies 3-D objects by looking for their notable features. This technique makes use of an evidence rule base, which is a set of salient or evidence conditions with corresponding evidence weights for various objects in the database. A measure of similarity between the set of observed features and the set of evidence conditions for a given object in the database is used to determine the identity of an object in the scene or reject the object(s) in the scene as unknown. This procedure has polynomial time complexity and correctly identifies a variety of objects in both synthetic and real range images. A technique for automatically deriving the evidence rule base from training views of objects is shown to generate evidence conditions that successfully identify new views of those objects     

Movie scene segmentation using background information

Scene extraction is the first step toward semantic understanding of a video. It also provides improved browsing and retrieval facilities to users of video database. This paper presents an effective approach to movie scene extraction based on the analysis of background images. Our approach exploits the fact that shots belonging to one particular scene often have similar backgrounds. Although part of the video frame is covered by foreground objects, the background scene can still be reconstructed by a mosaic technique. The proposed scene extraction algorithm consists of two main components: determination of the shot similarity measure and a shot grouping process. In our approach, several low-level visual features are integrated to compute the similarity measure between two shots. On the other hand, the rules of film-making are used to guide the shot grouping process. Experimental results show that our approach is promising and outperforms some existing techniques.     

Generating compact representations of static scenes by means of 3D object hierarchies

This paper presents a new heuristic algorithm for computing a compact hierarchical representation of the objects contained in a 3D static scene. The algorithm first builds a fully-connected adjacency graph that keeps the costs of grouping the different pairs of objects of the scene. Afterward, the graph’s minimum spanning tree is computed and its edges sorted in ascending order according to their cost. Next, from that sorted list, a cost-based clustering technique is applied, thus generating new objects at a higher level in the hierarchy. A new object can be defined after merging two or more objects according to their corresponding linking costs. The algorithm starts over by generating a new adjacency graph from those new objects, along with the objects that could not be merged before. The iterative process is applied until an adjacency graph with a single object is obtained. The latter is the root of the hierarchical representation. Balance and coherence of the hierarchy, in which spatially close objects are also structurally close, is achieved by defining an appropriate cost function. The proposed technique is evaluated upon several 3D scenes and compared to a previous technique. In addition, the benefits of the proposed technique with respect to techniques based on octrees and kd-trees are analyzed in terms of a practical application.     

Efficient rule mining and compression for RDF style KB based on Horn rules

The volume of published linked open datasets in RDF format is growing exponentially in the last decades. With this continuous proliferation of this growth, demands for managing, accessing, and compressing the RDF dataset have become increasingly important. Most approaches are focused on the structured compression technique while a very few researches have been done for compact representation of the RDF dataset. In this paper, we have proposed an efficient rule mining and compression approach for RDF datasets through various meaningful semantic association rules determined from the RDF graph. We have introduced grammar-based pattern system, clustering of rules, rules pruning, and Top-k scheme to improve the expressiveness of rule patterns, identify the similarity within the random pair of rules, extract the most delicate rules, find the accurate mining threshold, and efficiently learn the rules during the rule mining process from RDF Knowledge Base. Our proposed system uses Horn rules to achieve better compression through storing the triples matched with the precedent part while deleting the triples matched with the head part of the rules. For decreasing the mining time, we have introduced the ranking of the rules. The experimental result on the benchmark dataset asserts that our proposed rule mining and compression scheme has achieved approximately 22.10%, 40.5%, and 44% better compression than the exiting AMIE+, Rule-based compression, and TripleBit approaches, respectively. Our system also has achieved better performance both in terms of compression time and rule mining cost.

     

A neural network approach to CSG-based 3-D object recognition

Describes the recognition subsystem of a computer vision system based on constructive solid geometry (CSG) representation scheme. Instead of using the conventional CSG trees to represent objects, the proposed system uses an equivalent representation scheme-precedence graphs-for object representation. Each node in the graph represents a primitive volume and each are between two nodes represents the relation between them. Object recognition is achieved by matching the scene precedence graph to the model precedence graph. A constraint satisfaction network is proposed to implement the matching process. The energy function associated with the network is used to enforce the matching constraints including match validity, primitive similarity, precedence graph preservation, and geometric structure preservation. The energy level is at its minimum only when the optimal match is reached. Experimental results on several range images are presented to demonstrate the proposed approach     

Ray-driven dynamic working set rendering

Ray tracing a volume scene graph composed of multiple point-based volume objects (PBVO) can produce high quality images with effects such as shadows and constructive operations. A naive approach, however, would demand an overwhelming amount of memory to accommodate all point datasets and their associated control structures such as octrees. This paper describes an out-of-core approach for rendering such a scene graph in a scalable manner. In order to address the difficulty in pre-determining the order of data caching, we introduce a technique based on a dynamic, in-core working set. We present a ray-driven algorithm for predicting the working set automatically. This allows both the data and the control structures required for ray tracing to be dynamically pre-fetched according to access patterns determined based on captured knowledge of ray-data intersection. We have conducted a series of experiments on the scalability of the technique using working sets and datasets of different sizes. With the aid of both qualitative and quantitative analysis, we demonstrate that this approach allows the rendering of multiple large PBVOs in a volume scene graph to be performed on desktop computers.     

Permutation-test-based clustering method for detection of dynamic patterns in Spatio-temporal datasets

Massive spatio-temporal data have been collected from the earth observation systems for monitoring the changes of natural resources and environment. To find the interesting dynamic patterns embedded in spatio-temporal data, there is an urgent need for detecting spatio-temporal clusters formed by objects with similar attribute values occurring together across space and time. Among different clustering methods, the density-based methods are widely used to detect such spatio-temporal clusters because they are effective for finding arbitrarily shaped clusters and rely on less priori knowledge (e.g. the cluster number). However, a series of user-specified parameters is required to identify high-density objects and to determine cluster significance. In practice, it is difficult for users to determine the optimal clustering parameters; therefore, existing density-based clustering methods typically exhibit unstable performance. To overcome these limitations, a novel density-based spatio-temporal clustering method based on permutation tests is developed in this paper. High-density objects and cluster significance are determined based on statistical information on the dataset. First, the density of each object is defined based on the local variance and a fast permutation test is conducted to identify high-density objects. Then, a proposed two-stage grouping strategy is implemented to group high-density objects and their neighbors; hence, spatio-temporal clusters are formed by minimizing the inhomogeneity increase. Finally, another newly developed permutation test is conducted to evaluate the cluster significance based on the cluster member permutation. Experiments on both simulated and meteorological datasets show that the proposed method exhibits superior performance to two state-of-the-art clustering methods, i.e., ST-DBSCAN and ST-OPTICS. The proposed method can not only identify inherent cluster patterns in spatio-temporal datasets, but also greatly alleviates the difficulty in selecting appropriate clustering parameters.     

一种移动物体时空轨迹聚类的相似性度量方法

针对利用最小包围盒(MBB)压缩的移动物体时空轨迹,为了能对其进行有效地聚类,提出了一个基于盒内数据点密度的轨迹间相似性度量公式.首先,把两条轨迹的相似性度量转化为两条轨迹上有时间交叠的MBB之间的相似性度量,这在很大程度上减少了数据存储量.其次,分析两条轨迹上有时间交叠的MBB之间影响相似性的因素:时间持续、空间距离和盒内数据点的密度.剖析这3个因素对轨迹相似性的影响作用,提出了利用MBB压缩的移动物体时空轨迹相似性度量公式.实验证明采用本公式对移动物体时空轨迹进行聚类,可以提高聚类结果有效性指标Dunn的值.     

一种基于Kalman滤波的视频对象跟踪方法

为了更加准确地预测对象的位置和运动，利用刚体运动模型导出最佳Kalman系数，通过Kalman反馈滤波器对Moscheni等人提出的视频对象分割与跟踪算法进行改进，提出了一种将离散Kalman滤波技术用于视频序列的对象跟踪方法。这种方法可用于有关场景描述的各种应用领域中，如在机器视觉的研究中，对动态场景进行分析与理解；在基于对象的视频编码中（如MPEG-4），对视频对象进行分割后，分别进行编码，从而改善编码的可分级性及编码效率。实验结果表明，采用这种方法可以有效地改善时间-空间分割和目标跟踪，有助于更好地理解动态场景，并表现出良好的鲁棒性。     

Structure guided interior scene synthesis via graph matching

We present a method for reshuffle-based 3D interior scene synthesis guided by scene structures. Given several 3D scenes, we form each 3D scene as a structure graph associated with a relationship set. Considering both the object similarity and relation similarity, we then establish a furniture-object-based matching between scene pairs via graph matching. Such a matching allows us to merge the structure graphs into a unified structure, i.e., Augmented Graph (AG). Guided by the AG, we perform scene synthesis by reshuffling objects through three simple operations, i.e., replacing, growing and transfer. A synthesis compatibility measure considering the environment of the furniture objects is also introduced to filter out poor-quality results. We show that our method is able to generate high-quality scene variations and outperforms the state of the art.     

面向交通流量预测的时空超关系图卷积网络

交通流量预测是智能交通系统中的重要研究课题,然而,交通对象（如站点、传感器）之间存在的复杂局部时空关系使得这项研究颇具挑战。尽管以往的一些研究将流量预测问题转化为一个时空图预测问题从而取得了较大的进展,但是它们忽略了交通对象们跨时空维度的直接关联性。目前仍缺乏一种全面建模局部时空关系的方法。针对这一问题,首先提出一种新颖的时空超图建模方案,通过构造一种时空超关系来全面地建模复杂的局部时空关系;然后提出一种时空超关系图卷积网络（STHGCN）预测模型来捕获这些关系用于交通流量预测。在四个公开交通数据集上进行了大量对比实验,结果表明,相比ASTGCN、时空同步图卷积网络（STSGCN）等时空预测模型,STHGCN在均方根误差（RMSE）、平均绝对误差（MAE）、平均绝对百分比误差（MAPE）这三个评价指标上均取得了更优的结果,不同模型运行时间的对比结果也表明,STHGCN有着更高的推理速度。     

基于二分图的故障规则匹配优化算法

针对传统的基于相似度的故障规则匹配方法中未考虑输入条件与规则前件的整体匹配程度问题,采用二分图最优匹配方法对匹配过程进行优化,提出一种基于二分图的故障规则匹配优化算法,并将其应用于故障诊断推理．实例分析表明,与其他相似度匹配算法相比,所提出的方法有效提高了规则匹配的准确率,而且降低了时间消耗．     

A Similarity Metric for Retrieval of Compressed Objects: Application for Mining Satellite Image Time Series

This paper addresses the problem of building an index of compressed object databases. We introduce an informational similarity measure based on the coding length of two part codes. Then, we present a methodology for compressing the database by taking into account interobject redundancies and by using the informational similarity measure. The method produces an index included in the code of the data volume. This index is built such that it contains the minimal sufficient information to discriminate the data-volume objects. Then, we present an optimal two-part coder for compressing spatio-temporal events contained in satellite image time series (SITS). The two-part coder allows us to measure similarity and then to derive an optimal index of SITS spatio-temporal events. The resulting index is representative of the SITS information content and enables queries based on information content.     

一种基于多层语义特征的图像理解方法

视觉场景理解包括检测和识别物体、推理被检测物体之间的视觉关系以及使用语句描述图像区域.为了实现对场景图像更全面、更准确的理解,将物体检测、视觉关系检测和图像描述视为场景理解中3种不同语义层次的视觉任务,提出一种基于多层语义特征的图像理解模型,并将这3种不同语义层进行相互连接以共同解决场景理解任务.该模型通过一个信息传递图将物体、关系短语和图像描述的语义特征同时进行迭代和更新,更新后的语义特征被用于分类物体和视觉关系、生成场景图和描述,并引入融合注意力机制以提升描述的准确性.在视觉基因组和COCO数据集上的实验结果表明,所提出的方法在场景图生成和图像描述任务上拥有比现有方法更好的性能.     

Crowd flux analysis and abnormal event detection in unstructured and structured scenes

This paper presents a spatio-temporal grid-based framework to deal with the complexity of structured and unstructured motion flows that can effectively group optical flows in the field of view into crowds. This approach utilizes motion flows of the features based on a grid in a scene. In order to detect abnormal events in crowded scenes, the proposed method measures motion features including the speed and direction of moving objects based on a spatio-temporal grid-based approach for flow representation. Experiments have been conducted on several different videos in three domains that are crosswalks, escalators, and highways. To evaluate and compare the performance of our method to other methods, ROC curves are plotted which take into consideration both detection rate and false alarm rate for multiple threshold values.     

通过语义控制改进运动合成

运动相似性度量是基于运动图的人体运动合成技术的关键.现有方法主要使用运动捕捉数据姿态数值特征进行相似性度量与合成控制,很难处理语义描述上同类型运动数据集中的不同运动实现版本间的时间与空间特征分布变化的问题.提出了引入用户语义控制来改进基于运动图的人体运动合成方法.使用关系特征作为高层语义描述,刻画同类型运动的空间变化特征;通过自学习过程,获得运动类模板,作为同类型运动的空-时特征语义描述;通过将运动类模板数据与人体运动序列文件进行匹配,实现运动类型识别和自动语义信息标注;借助关系特征语义描述及运动序列文件的语义标注信息,实现在基于运动图的运动合成中引入用户直观的高层语义控制.运动合成实验结果显示了该方法的有效性,为获得高质量人体运动合成数据提供途径.     

基于时空渐进特征模型的抗遮挡多目标跟踪

针对多目标跟踪过程中存在目标相互遮挡的问题,提出一种基于时空渐进特征模型的抗遮挡多目标跟踪方法.该方法根据目标的关联状态、重叠情况、深度顺序构建遮挡推理模型,在线识别被遮挡目标, 并对其进行基于时空渐进特征模型的目标再检测.首先,利用目标特征模型中的空间位置参数确定搜索区域;然后,计算目标模型与检测响应之间的表观特征相似度,引入一种自适应阈值,并选择相近程度较高的检测响应为候选量测;最后,采用时间特征匹配的方式获得目标真实检测响应,完成多目标跟踪的任务.通过MOT15和MOT17数据集中的实验结果表明,所提出方法能够有效地识别被遮挡目标,可在多种复杂场景下实现相互遮挡目标的鲁棒性跟踪,相对于现有方法具有较高的准确性、精度和轨迹完整性.     

A model of fuzzy spatio-temporal knowledge representation and reasoning based on high-level Petri nets

In many application areas there is a need to represent human-like knowledge related to spatio-temporal relations among multiple moving objects. This type of knowledge is usually imprecise, vague and fuzzy, while the reasoning about spatio-temporal relations is intuitive. In this paper we present a model of fuzzy spatio-temporal knowledge representation and reasoning based on high-level Petri nets. The model should be suitable for the design of a knowledge base for real-time, multi-agent-based intelligent systems that include expert or user human-like knowledge. The central part of the model is the knowledge representation scheme called FuSpaT, which supports the representation and reasoning for domains that include imprecise and fuzzy spatial, temporal and spatio-temporal relationships. The scheme is based on the high-level Petri nets called Petri nets with fuzzy spatio-temporal tokens (PeNeFuST). The FuSpaT scheme integrates the theory of the PeNeFuST and 117 spatio-temporal relations.The reasoning in the proposed model is a spatio-temporal data-driven process based on the dynamical properties of the scheme, i.e., the execution of the Petri nets with fuzzy spatio-temporal tokens. An illustrative example of the spatio-temporal reasoning for two agents in a simplified robot-soccer scene is given.     

Spatio-temporal composition of video objects: representation andquerying in video database systems

A key characteristic of video data is the associated spatial and temporal semantics. It is important that a video model models the characteristics of objects and their relationships in time and space. J.F. Allen's (1983) 13 temporal relationships are often used in formulating queries that contain the temporal relationships among video frames. For the spatial relationships, most of the approaches are based on projecting objects on a two or three-dimensional coordinate system. However, very few attempts have been made formally to represent the spatio-temporal relationships of objects contained in the video data and to formulate queries with spatio-temporal constraints. The purpose of the work is to design a model representation for the specification of the spatio-temporal relationships among objects in video sequences. The model describes the spatial relationships among objects for each frame in a given video scene and the temporal relationships (for this frame) of the temporal intervals measuring the duration of these spatial relationships. It also models the temporal composition of an object, which reflects the evolution of object's spatial relationships over the subsequent frames in the video scene and in the entire video sequence. Our model representation also provides an effective and expressive way for the complete and precise specification of distances among objects in digital video. This model is a basis for the annotation of raw video