Space Subdivision Method Research Articles

Indoor functional subspace division from point clouds based on graph neural network

Indoor scenes are closely related to human life and contain rich geometric and semantic information. Dividing indoor spaces from data is crucial for multiple applications, such as navigation and digital twins. However, achieving this task is challenging. Traditional indoor space division methods only represent buildings, floors, and rooms to some extent, lacking semantic descriptions of indoor elements and their spatial arrangements. To divide an indoor space more finely, a novel indoor space subdivision method is proposed in this paper. Our method leverages the flexible space subdivision framework (FSS) to categorize indoor space into free navigation subspace, object subspace, and functional subspace. To define functional subspaces, we present a taxonomy for the spatial layout patterns of common indoor elements (like tables, chairs, ceilings, walls, floors, etc.). Then, we introduce scene graphs to represent indoor 3D scenes, where each node represents an indoor element and each edge encodes the spatial relationship between the elements. Finally, a node classification network is proposed to segment indoor scene into subspaces and predicts their (semantic) functions. We select 9 buildings of Matterport3D and 6 areas in S3DIS and merge them to form our dataset for training and testing our method. Experiments yield good results with up to 90.42% accuracy and 85.28% F1-scores in overall space subdivision. Moreover, compared with the various graph node classification networks, our method has achieved the best performance in indoor space subdivisions.

Dynamic optimal space partitioning for redirected walking in multi-user environment

In multi-user Redirected Walking (RDW), the space subdivision method divides a shared physical space into sub-spaces and allocates a sub-space to each user. While this approach has the advantage of precluding any collisions between users, the conventional space subdivision method suffers from frequent boundary resets due to the reduction of available space per user. To address this challenge, in this study, we propose a space subdivision method called Optimal Space Partitioning (OSP) that dynamically divides the shared physical space in real-time. By exploiting spatial information of the physical and virtual environment, OSP predicts the movement of users and divides the shared physical space into optimal sub-spaces separated with shutters. Our OSP framework is trained using deep reinforcement learning to allocate optimal sub-space to each user and provide optimal steering. Our experiments demonstrate that OSP provides higher sense of immersion to users by minimizing the total number of reset counts, while preserving the advantage of the existing space subdivision strategy: ensuring better safety to users by completely eliminating the possibility of any collisions between users beforehand. Our project is available at https://github.com/AppleParfait/OSP-Archive.

Space Subdivision in Indoor Mobile Laser Scanning Point Clouds Based on Scanline Analysis.

Indoor space subdivision is an important aspect of scene analysis that provides essential information for many applications, such as indoor navigation and evacuation route planning. Until now, most proposed scene understanding algorithms have been based on whole point clouds, which has led to complicated operations, high computational loads and low processing speed. This paper presents novel methods to efficiently extract the location of openings (e.g., doors and windows) and to subdivide space by analyzing scanlines. An opening detection method is demonstrated that analyses the local geometric regularity in scanlines to refine the extracted opening. Moreover, a space subdivision method based on the extracted openings and the scanning system trajectory is described. Finally, the opening detection and space subdivision results are saved as point cloud labels which will be used for further investigations. The method has been tested on a real dataset collected by ZEB-REVO. The experimental results validate the completeness and correctness of the proposed method for different indoor environment and scanning paths.

Event driven algorithms applied to a high energy ball mill simulation

In the physics community, event driven molecular dynamics simulations have been successfully applied to typical physical topics, such as granular gases. In contrast, in the engineering sciences, event driven algorithms have not become as popular, even though they can be far more efficient for special systems. In this paper, we present an event driven simulation of a high energy ball mill, where the motion and energy transfer of the grinding balls during operation are simulated. Stable numerical algorithms for collision detection and collision response are developed and several possible pitfalls are discussed. Furthermore, an improved event list handling technique and a specialized space subdivision method is presented. The performance of the partitioning is demonstrated by experiments. It is shown that the scaling rules usually applied are oversimplified. A new way of scaling the process parameters to obtain a higher production yield using iso-impact diagrams is presented.

Novel Techniques for Robust Voxelization and Visualization of Implicit Surfaces

Voxelization is the transformation of geometric surfaces into voxels. Up to date this process has been done essentially using incremental algorithms. Incremental algorithms have the reputation of being efficient but they lack an important property: robustness. The voxelized representation should envelop its continuous model. However, without robust methods this cannot be guaranteed. This article describes novel techniques of robust voxelization and visualization of implicit surfaces. First of all our recursive subdivision voxelization algorithm is reviewed. This algorithm was initially inspired by Duff's image space subdivision method. Then, we explain the algorithm to voxelize implicit surfaces defined in spherical or cylindrical coordinates. Next, we show a new technique to produce infinite replications of implicit objects and their voxelization method. Afterward, we comment on the parallelization of our voxelization procedure. Finally we present our voxel visualization algorithm based on point display. Our voxelization algorithms can be used with any data structure, thanks to the fact that a voxel is only stored once the last subdivision level is reached. We emphasize the use of the octree, though, because it is a convenient way to store the discrete model hierarchically. In a hierarchy the discrete model refinement is simple and possible from any previous voxelized scene thanks to the fact that the voxelization algorithms are robust.

A new space subdivision for ray tracing CSG solids

Ray tracing successfully creates realistic images of constructive solid geometry (CSG) solids. We describe a nonuniform space subdivision scheme that reduces both the number of ray-object intersection computations and point classifications. Our method uses the face planes of the primitives' S-bounds in a bottom-up fashion and produces a subdivision wherein the localized CSG tree in each leaf voxel is greatly minimized. The use of S-bounds in the space subdivision effectively reduces the number of intersection computations as well. The reduction of the localized CSG tree in turn further reduces the number of intersection computations and point classifications. We briefly review existing methods for ray tracing CSG solids, describe our proposed space subdivision method, discuss our implementation and compare it to Bouatouch's (1987) method, and summarize our test results. >

Volume occupation, environment and accessibility in proteins. The problem of the protein surface

The volumes occupied by atoms or groups of atoms in ribonuclease S are calculated, using a new treatment of the surface, and considering buried and exposed atoms separately. The volumes occupied by particular atom types fall within a narrow range, with standard deviations between 10% and 15% of the mean. Summation of volumes over main and side-chain groups removes inaccuracies in the single atom data, and results in even narrower distributions. Comparison of the volume spreads with those found in model glasses suggests that the remaining variation is unlikely to be much reduced by improving further either the space-subdivision method or the surface treatment. These improved volume distributions might be used as more stringent criteria against which to test trial tertiary structures. The computer program used produces additional information that could be used to investigate quantitatively the possible freedom of movement in different parts of the molecule. It also provides complementary data on environment and surface exposure.