Mechanical, Electrical and Plumbing (MEP) models are generally characterized by information redundancy and a high density of irregularly shaped components. Consequently, they require large storage spaces and are not conducive for... more
Mechanical, Electrical and Plumbing (MEP) models are generally characterized by information redundancy and a high density of irregularly shaped components. Consequently, they require large storage spaces and are not conducive for interchange purposes. Geometric optimization of MEP models can play a significant role in facilitating model exchange and handover by increasing storage, transmission and display efficiency. To date, the body of knowledge on such geometric optimization, unfortunately, is still very narrow.
This paper presents and tests a solution for the optimization of storage, transmission, and display of MEP models. Storage optimization was achieved through a mapping-based model description method and a novel Quadric-Error-Metric (QEM) mesh simplification algorithm, reducing required storage while maintaining the contour of components. For transmission optimization, a normal vector compression algorithm and a fixed-dictionary specific compression algorithm were proposed to achieve efficient compression of data thus, fulfilling the need for cross-platform interchange. Display optimization was obtained through a normal vector regeneration algorithm for clustering triangle meshes to improve 3D display effects.
The evaluation of the solution, performed for each individual component separately as well as for an entire solution, proved successful. Volume of storage data was reduced by 40% to 50% through mesh simplification. Data transmission volume was reduced by more than 80% for components with complicated geometry without affecting the topology of the components. Finally, the display process was capable of decreasing the number of triangles and delivering very good quality of displayed models.
In this paper, we introduce a novel reconstruction and modeling pipeline to create polygonal models from unstructured point clouds. We propose an automatic polygonal reconstruction that can then be interactively refined by the user. An... more
In this paper, we introduce a novel reconstruction and modeling pipeline to create polygonal models from unstructured point clouds. We propose an automatic polygonal reconstruction that can then be interactively refined by the user. An initial model is automatically created by extracting a set of RANSAC-based locally fitted planar primitives along with their boundary polygons, and then searching for local adjacency relations among parts of the polygons. The extracted set of adjacency relations is enforced to snap polygon elements together, while simultaneously fitting to the input point cloud and ensuring the planarity of the polygons. This optimization-based snapping algorithm may also be interleaved with user interaction. This allows the user to sketch modifications with coarse and loose 2D strokes, as the exact alignment of the polygons is automatically performed by the snapping. The generated models are coarse, offer simple editing possibilities by design and are suitable for interactive 3D applications like games, virtual environments etc. The main innovation in our approach lies in the tight coupling between interactive input and automatic optimization, as well as in an algorithm that robustly discovers the set of adjacency relations.
