6.1 Conclusion
In this thesis, we introduce our novel method for ice melting simulation with render-ing. Our method includes three main contributions as shown below:
1. We give an attribute for each ice particle named virtual water volume. Virtual water volume will be transferred between ice particles. We use this method to simulate water flowing on the surface of ice model.
2. We use a new method to calculate the potential field and achieve smoothly shrinking ice model. The latent heat of ice particles and virtual water volume are taken into account for computing the potential field.
3. A ray tracing method is used to render meshes and metaballs in the same scene.
We propose a special principles of judgment to find intersections between rays and objects (i.e., meshes or metaballs).
Our ice melting method is based on the thermal energy transfer between particles, and smoothed particle hydrodynamics (SPH) is used on fluid simulation. The ice particles are decided by voxelizing the input polygonal model. Heat transfer occurs between particles, from surrounding air to ice particles, and from the heat source to particles. The ice particles are removed if they absorb enough heat.
We use virtual water volume transfer to simulate the flowing water on the surface of ice model. Virtual water volume of an ice particle is increased by increasing the latent heat of the ice particle. The water particle is generated when the amount of water volume in an ice particle is too large. The motion of water particles is based on SPH. Water particles
are absorbed by ice particles and transfer to virtual water volume if the water particles is located on the top of ice particles.
An improvement of computing the potential field for the marching cubes is proposed.
The shape of the ice model will be affected by the latent heat and virtual water volume through the field function. Therefore, the shape of ice model will shrink smoothly.
In the rendering procedure, the ice model and water particles are separated and we use a mixed method to render them in the same scene. We use linear bounding volume hierarchy (LBVH) to create hierarchy constructions for the ice model and metaballs re-spectively. Then, we use the ray tracing method with the special principles of judgment to find the intersection for the rays.
The experimental results show that our algorithm of ice melting simulation can achieve more realistic animation. The previous method [IUDN10] will directly transform the ice particle into water particle if the ice particle absorbs enough energy. However, their result is not realistic. Water should flow on the surface. Our method can handle this phenomenon by transferring virtual water volume between ice particles. The execution performance of our rendering procedure is acceptable, and our result is more realistic than the other meth-ods.
6.2 Future Work
In the real world, the environment forces should affect the motion of the ice model.
For example, part of icicle is separated and falls down to the floor because of gravity, or ice cubes float in the water. In our system, we skip these forces mainly because the motion of an ice model will change the result of marching cubes. The shape of the ice model deforms violently when the ice model moves fast or rotates between two frames.
We want to handle this situation in the future.
Water particles moving in the scene may pierce the surface of the ice model (i.e.
there are overlapping portions between water particles and the ice model). ice model is a polygonal model which is made by marching cubes method. We don't calculate the collision detection for the water particles and ice model. Instead, we calculate the pressure defined in SPH method between the ice particles and water particles, so the water particles maintain appropriate distance to the ice model. However, piercing between ice and water particles happens when the force of water particles are too large. We will add the collision detection and response in our system.
Our method uses the latent heat to control the marching cubes. The ice model will
ear. The meshes of these places shrink too fast and disappear before the ice particles absorb enough heat. Therefore, the water particle will unexpectedly appear in the air, as shown in Figure 6.1. We need to handle this situation by giving a better adjustment for field value.
Figure 6.1: The sketch map of problem in meshes shrinking.
Finally, some ice characteristic in our simulation are not handled, such as bubbles or cracks in the ice model. In the rendering part, we haven't considered complex lighting, such as focus, and defocus. In the future, we will find some approaches to implement these characteristics in our simulation.
Bibliography
[App68] A. Appel. Some techniques for shading machine renderings of solids. In Proceedings of the April 30--May 2, 1968, spring joint computer conference, pages 37--45, 1968.
[AW87] J. Amanatides and A Woo. A fast voxel traversal algorithm for ray tracing.
Eurographics, 87:3--10, 1987.
[CHH02] N.A. Carr, J.D. Hall, and J.C. Hart. The ray engine. In Proceedings of the ACM SIGGRAPH/EUROGRAPHICS conference on Graphics hardware, pages 37--46, 2002.
[CMJG07] Wojtan C., Carlson M., Mucha P. J., and Turk G. Animating corrosion and erosion. In EurographicsWorkshop on Natural Phenomena, pages 15--22, 2007.
[FA01] I. Fujishiro and E. Aoki. Volume graphics modeling of ice thawing. Volume Graphics, pages 69--80, 2001.
[FGFR06] Losasso F., Irving G., Guendelman F., and Fedkiw R. Melting and burning solids into liquids and gases. In IEEE Transactions on Visualization and Computer Graphics, volume 12, pages 343--352, 2006.
[FM07] Makoto Fujisawa and Kenjiro T. Miura. Animation of ice melting phe-nomenon based on thermodynamics with thermal radiation. In GRAPHITE, pages 249--256, 2007.
[FS05] T. Foley and J. Sugerman. KD-tree acceleration structures for a GPU ray-tracer. In Proceedings of the ACM SIGGRAPH/EUROGRAPHICS confer-ence on Graphics hardware, pages 15--22, 2005.
[GPP+10] O. Gourmel, A. Pajot, M. Paulin, L. Barthe, and P Poulin. Fitted BVH for Fast Raytracing of Metaballs. In Computer Graphics Forum, volume 29, page 281–288, 2010.
[HSHH07] D.R. Horn, J. Sugerman, M. Houston, and P. Hanrahan. Interactive kd tree GPU raytracing. In Proceedings of the 2007 symposium on Interactive 3D graphics and games, page 174, 2007.
[IUDN10] K. Iwasaki, H. Uchida, Y. Dobashi, and T. Nishita. Fast Particle-based Visual Simulation of Ice Melting. In Computer Graphics Forum, volume 29, pages 2215--2223, 2010.
[Kau87] A. Kaufman. An algorithm for 3D scan-conversion of polygons. Eurograph-ics, pages 197--208, 1987.
[Khr09] Khronos OpenCL Working Group. OpenCL 1.0 Specification, 2009.
[KSN08] Y. Kanamori, Z. Szego, and T. Nishita. GPU-based Fast Ray Casting for a Large Number of Metaballs. In Computer Graphics Forum, volume 27, pages 351--360, 2008.
[LGS+09] C. Lauterbach, M. Garland, S. Sengupta, D. Luebke, and D. Manocha. Fast BVH construction on GPUs. In Computer Graphics Forum, volume 28, pages 375--384, 2009.
[NN94] T. Nishita and E Nakamae. A method for displaying metaballs by using Bézier clipping. Computer Graphics Forum, 13:271--280, 1994.
[NVI] NVIDIA Corporation. GPU computing SDK, http://developer.nvidia.com/
gpu-computing-sdk.
[PBMH02] T. J. Purcell, I. Buck, W. R. Mark, and P Hanrahan. Ray tracing on pro-grammable graphics hardware. volume 21, pages 703--712, 2002.
[PGSS07] S. Popov, J. Günther, H.P. Seidel, and P. Slusallek. Stackless kd-tree traver-sal for high performance GPU ray tracing. In Computer Graphics Forum, volume 26, pages 415--424, 2007.
[PL10] J. Pantaleoni and D. Luebke. HLBVH: hierarchical LBVH construction for real-time ray tracing of dynamic geometry. HPG '10, pages 87--95, 2010.
[SEL11] Erik Smistad, Anne C. Elster, and Frank Lindseth. Fast surface extraction and visualization of medical images using OpenCL and GPUs. In MICCAI, 2011.
[SSP07] B. Solenthaler, J. Schläfli, and R. Pajarola. A unified particle model for fluid–solid interactions. Comp. Anim. Virtual Worlds, 18:69--82, 2007.
[Whi80] Turner Whitted. An improved illumination model for shaded display. Com-mun. ACM, 23(6):343--349, 1980.
[WMG+07] I. Wald, W.R. Mark, J. Günther, S. Boulos, T. Ize, W. Hunt, S.G. Parker, and P. Shirley. State of the art in ray tracing animated scenes. Eurographics 2007 State of the Art Reports, pages 89--116, 2007.
[ZHWG08] K. Zhou, Q. Hou, R. Wang, and B. Guo. Real-time kd-tree construction on graphics hardware. In ACM SIGGRAPH Asia 2008 papers, pages 1--11, 2008.
[ZY06] Qiu F. Kaufman A. Mueller K. Zhao Y., Wang L. Melting and flowing in multiphase environment. In Compute and Graphics, volume 30, pages 519--528, 2006.