Efficient polygonization of tree trunks modeled by convolution surfaces

Abstract

We present an efficient polygonization approach for tree trunks modeled by line skeleton-based convolution surfaces. A quad-dominated non-convex bounding polyhedron is firstly created along the skeleton, which is then tetrahedralized and subdivided into the pre-defined resolution. After that, the iso-surface within each tetrahedron is extracted using marching tetrahedra. Our algorithm can generate polygons with adaptive edge lengths according to the thickness of the trunk. In addition, we present an efficient CUDA-based parallel algorithm utilizing the high parallelism of the tetrahedron subdivision, the potential field calculation, and the iso-surface extraction.

High-quality tree structures modelling using local convolution surface approximation

“…In order to extract the iso-surface of the tree trunk-based convolution surface, the improved Marching tetrahedra [59] (Fig. 11(a)) produces triangles efficiently, which has better visual effects and achieves higher performance than the traditional Marching tetrahedra.…”

“…Then the line skeletons are extracted using the same approach as in [59], where an implicit Laplacian smoothing operator is involved [6,13]. Besides the extracted skeletons, a useful by-product of induced skeleton-mapping is generated.…”

“…Furthermore, it is prone to missing small twigs for complex tree models because the output of the Marching cubes is resolutiondependent. Even though there are a large number of improvements [55,8,10,1,40,59], it is still difficult to balance the quality of the iso-surface polygons and the performance. To solve these problems, we propose an interactive GPU-based quad-only tessellation method to polygonize convolution surfaces with good edge flows along the skeletons of tree models.…”

Efficient tree-like structures modeling based on subdivision surfaces

“…The analytical solutions for t m .m D 1; 2; 3; : : : ; 1/ are given as t m D 5184 p 2 32 .149 C 13 .m/ C 101 .m// (7) which controls the approximation precision in the skeletal direction ( Figure 5). In our experiments, 8 < :…”

“…Among various existing approaches of surface reconstruction from skeletal data, the convolution surface-based approach [5][6][7] generates a smooth surface but involves huge potential field computation and iso-surface extraction. In addition, it usually generates low-quality triangle meshes.…”