The problem of tracking multiple objects has been investigated in various research and industrial fields. Among existing methods, random finite set (RFS) solutions such as the generalized labeled ...multi-Bernoulli (GLMB) filter has provided efficient solutions with solid theoretical justifications. Furthermore, implementations show that the GLMB approach is efficient under challenging scenarios. In this paper, we study an RFS-based method for multi-object tracking (MOT) through a simple data structure for label partitioning. Specifically, grid index structure based techniques for splitting a label space and a label-partitioned GLMB tracker are investigated. We finally evaluate the performance of label partitioning and the GLMB filter via various means such as visualization, execution time, and MOT metrics.
A framework for interactive modeling of three dimensional caves is presented. It is based on a new spatial data structure that extends existing terrain rendering methods. The cave is represented as a ...set of slabs, each encoding a portion of the model along an axis. We describe methods for a user to modify this model locally and procedural methods for global alteration. We wish to allow cave modeling as easily as existing terrain editing programs that restrict the model to a single two dimensional manifold. In this paper, we discuss existing cave visualization programs, including their limitations, as well as how terrain editing and rendering methods can be used in the process of modeling caves.
A hybrid 3D spatial data structure for the integration of aboveground, ground and underground objects based on spatial cognition theory is proposed, which considers the space subdivision and topology ...completeness. Object-oriented method is used to abstract objects in real world which is defined into three levels. Geometry primitives and topology primitives are defined in this data structure. Topology is the option of the geometry. The geometry information can be coded in a binary method to store in the relation database. An experiment is designed to integrate the buildings, terrain, and geological stratification seamlessly. The data structure could express aboveground, ground and underground objects effectively.
The usual quadtree node non-pointer codification is based on interleaved binary representations of node coordinates, in such a way that every operation that concerns to the spatial position or to the ...specific orientation of the region represented by the node needs to undo this interleaving process. So, the computation time of such operations is linear with the node depth. In this paper an alternative codification is presented called “non-interleaved codification”. The new codification has a simpler management and a higher intuitiveness than current codifications that use the interleaving approach. The proposed codification is more efficient than previous ones for the following set of operations: generating the node codes from the spatial coordinates, recovering original coordinates from the node codes, and performing topological operations where explicit or implicit reference is made to node location, for instance, checking if two nodes are adjacent, evaluating distances between nodes, evaluating relative orientation, etc. The proposed codification performs all these operations in O (1) time, independently from the node depth.
In order to meet the requirements of large-scale 3D scene transmission, parsing and real-time visualization in the Web environment, a kind of data structure is proposed in this paper, which specifies ...the logical structure and physical storage format of the 3D spatial data, and can effectively divide and index the 3D scene. Meanwhile, based on the existing spatial index technology, we propose a hybrid index structure R* O-tree which combines the R* -tree and the octree to improve the performance. The experimental results show that the proposed data structure can save about 60%-80% disk space, and the improved index structure is more efficient. Our method is more advantageous than the traditional three-dimensional spatial data structure.
We propose a novel data structure, called the DIMD-tree, that provides efficient successive spatial searches in GPS navigation systems, mobile phone systems and so forth. The DIMD-tree can be ...applicable for the geographic data management of positioning systems as well as the query processing for current and anticipated future positions of moving objects. The basic idea introduced in the DIMD-tree is direct indexing into a node or leaf which includes the predicted position of a moving object. The performances of the DIMD-tree are evaluated by simulation tests.
Utilizing the vector units of current processors for ray tracing single rays through Bounding Volume Hierarchies has been accomplished by increasing the branching factor of the acceleration structure ...to match the vector width. A high branching factor allows vectorized bounding box tests but requires a complex control flow for the calculation of a front-to-back traversal order. We propose a novel algorithm for single rays entirely based on vector operations that performs a complete traversal iteration in constant time, ideally suited for current and future micro architectures featuring wide vector units. In addition we use our single ray technique as a building block to construct a fast packet traversal for coherent rays. We validate our algorithms with implementations utilizing the AVX2 and AVX-512 instruction sets and demonstrate significant performance gains over state-of-the-art solutions.
R-trees are an important spatial data structure for database storage. In this paper, we introduce a new method, called
branch grafting method, for R-tree implementation. The major motivation of this ...proposed technique is to produce more desirable result and to reduce the number of nodes created in the R-tree. We present this method by comparing it with reinsertion method as used in R*-tree. Experiments have been carried out, and the strength of this proposed method is outlined.
In the engineering database system, multiple versions of a design including engineering drawings should be managed efficiently. The paper proposes an extended spatial data structure for efficient ...management of multiversion engineering drawings. The R-tree is adapted as a basic data structure. The efficient mechanism to manage the difference between drawings is introduced to the R-tree to eliminate redundant duplications and to reduce the amount of storage required for the data structure. The extended data structures of the R-tree, MVR and MVR* trees, are developed and the performances of these trees are evaluated. A series of simulation tests shows that, compared with the basic R-tree, the amounts of storage required for the MVR and MVR* trees are reduced to 50% and 30%, respectively. The search efficiencies of the R, MVR, and MVR* trees are almost the same.
PDF
