액체 시뮬레이션의 얇은 특징을 빠르게 표현하기 위한 CPU와 GPU 이기종 컴퓨팅 기술 = A CPU and GPU Heterogeneous Computing Techniques for Fast Representation of Thin Features in Liquid Simulations

다국어 초록 (Multilingual Abstract)

We propose a new method particle-based method that explicitly preserves thin liquid sheets for animating liquids on CPU--GPU heterogeneous computing framework. Our primary contribution is a particle-based framework that splits at thin points and collapses at dense points to prevent the breakup of liquid on GPU. In contrast to existing surface tracking methods, the our method does not suffer from numerical diffusion or tangles, and robustly handles topology changes on CPU--GPU framework. The thin features are detected by examining stretches of distributions of neighboring particles by performing PCA(Principle component analysis), which is used to reconstruct thin surfaces with anisotropic kernels. The efficiency of the candidate position extraction process to calculate the position of the fluid particle was rapidly improved based on the CPU--GPU heterogeneous computing techniques. Proposed algorithm is intuitively implemented, easy to parallelize and capable of producing quickly detailed thin liquid animations.

국문 초록 (Abstract)

우리는 유체의 얇은 막을 명시적으로 표현하고 보존할 수 있는 CPU--GPU 이기종 컴퓨팅 기반의 유체 시뮬레이션 기법을 소개한다. 본 논문에서 가장 큰 기여는 얇은 유체표면에서 쪼개지거나 밀도가 높은 지점에서 붕괴되어 유체표면에 나타나는 Hole을 방지하는 입자 기반 프레임워크를 GPU를 활용한다는 것이다. 유체표면을 추적하는 기존의 방법과는 달리, 제안된 프레임워크는 CPU--GPU 프레임워크상에서 수치적 확산이나 꼬임문제 없이 안정적으로 토폴로지 변화를 처리할 수 있다. 얇은 표면의 특징은 이방성 커널(Anisotropic kernel)과 주성분 분석(Principal component analysis; PCA)을 GPU상에서 수행하여 유체의 방향성을 빠르게 찾고, 새로운 유체입자의 위치를 결정하기 위해 계산하는, 후보위치 추출 과정의 효율성을 CPU--GPU 이기종 컴퓨팅 기술 기반으로 빠르게 계산한다. 제안된 알고리즘은 직관적으로 구현되며, 병렬화가 쉽고 시각적으로 디테일한 액체의 얇은 표면을 빠르게 애니메이션 할 수 있다.

참고문헌 (Reference)

1 J.-M. Hong, "Wrinkled flames and cellular patterns" 26 (26): 2007

2 M. Becker, "Weakly compressible sph for free surface flows" 209-217, 2007

3 T. Kim, "Wavelet turbulence for fluid simulation" 27 (27): 50-, 2008

4 C. W. Hirt, "Volume of fluid (vof) method for the dynamics of free boundaries" 39 (39): 201-225, 1981

5 F. Losasso, "Twoway coupled sph and particle level set fluid simulation" 14 (14): 797-804, 2008

6 B. Yang, "Turbulence synthesis for shapecontrollable smoke animation" 25 (25): 465-472, 2014

7 J. Glimm, "Three-dimensional front tracking" 19 (19): 703-727, 1998

8 V. Mihalef, "Textured liquids based on the marker level set" 26 (26): 457-466, 2007

9 T. Pfaff, "Synthetic turbulence using artificial boundary layers" 121:1-121:10, 2009

10 D. Kim, "Stretching and wiggling liquids" 28 (28): 120:1-120:7, 2009

11 M. Kass, "Snakes: Active contour models" 1 (1): 321-331, 1988

12 T. Harada, "Smoothed particle hydrodynamics on gpus" 63-70, 2007

13 S. Lahabar, "Singular value decomposition on gpu using cuda" 1-10, 2009

14 F. Losasso, "Simulating water and smoke with an octree data structure" 23 (23): 457-462, 2004

15 T. Jang, "Simulating sph fluid with multi-level vorticity" 10 (10): 21-, 2011

16 T. Pfaff, "Scalable fluid simulation using anisotropic turbulence particles" 29 (29): 174-, 2010

17 J. Yu, "Reconstructing surfaces of particlebased fluids using anisotropic kernels" 32 (32): 5-, 2013

18 H. Yan, "Real-time fluid simulation with adaptive sph" 20 (20): 417-426, 2009

19 B. Solenthaler, "Predictive-corrective incompressible sph" 28 (28): 40-, 2009

20 S. Premzoe, "Particle-based simulation of fluids" 22 (22): 401-410, 2003

21 M. Muller, "Particle-based fluid simulation for interactive applications" 154-159, 2003

22 F. H. Harlow, "Numerical calculation of timedependent viscous incompressible flow of fluid with free surface" 8 (8): 2182-2189, 1965

23 M. Pauly, "Meshless animation of fracturing solids" 24 (24): 957-964, 2005

24 P. Goswami, "Interactive sph simulation and rendering on the gpu" 55-64, 2010

25 R. Li, "Gpu-accelerated preconditioned iterative linear solvers" 63 (63): 443-466, 2013

26 C. Dyken, "Gpu-accelerated data expansion for the marching cubes algorithm" 115-123, 2010

27 S. Osher, "Fronts propagating with curvature-dependent speed: algorithms based on hamiltonjacobi formulations" 79 (79): 12-49, 1988

28 B. M. Klingner, "Fluid animation with dynamic meshes" 25 (25): 820-825, 2006

29 N. Heo, "Detail-preserving fully-eulerian interface tracking framework" 29 (29): 176-, 2010

30 D. Kim, "Baroclinic turbulence with varying density and temperature" 18 (18): 1488-1495, 2012

31 D. Kim, "Baroclinic turbulence with varying density and temperature" 18 (18): 1488-1495, 2012

32 Y. Zhu, "Animating sand as a fluid" 24 (24): 965-972, 2005

33 Z. Wang, "An improved particle correction procedure for the particle level set method" 228 (228): 5819-5837, 2009

34 B. Adams, "Adaptively sampled particle fluids" 2007

35 A. Selle, "A vortex particle method for smoke, water and explosions" 24 (24): 910-914, 2005

36 R. Keiser, "A unified lagrangian approach to solid-fluid animation" 125-148, 2005

37 A.W. Bargteil, "A semi-lagrangian contouring method for fluid simulation" 25 (25): 19-38, 2006

38 F. Sin, "A point-based method for animating incompressible flow" 247-255, 2009

39 D. Gerszewski, "A point-based method for animating elastoplastic solids" 133-138, 2009

40 R. Ando, "A particle-based method for preserving fluid sheets" 7-16, 2011

41 S. Koshizuka, "A particle method for incompressible viscous flow with fluid fragmentation" 4 : 29-, 1995

42 H. Bhatacharya, "A level-set method for skinning animated particle data" 17-24, 2011

43 D. Enright, "A hybrid particle level set method for improved interface capturing" 183 (183): 83-116, 2002

44 J. F. Blinn, "A generalization of algebraic surface drawing" 1 (1): 235-256, 1982

45 G. Tryggvason, "A fronttracking method for the computations of multiphase flow" 169 (169): 708-759, 2001

46 C. Batty, "A fast variational framework for accurate solid-fluid coupling" 26 (26): 100-, 2007

47 Y. Dobashi, "A fast simulation method using overlapping grids for interactions between smoke and rigid objects" 27 (27): 477-486, 2008