Efficient Light Transport Algorithms for Computer Graphics

已经设计了许多逼真的图像合成算法，但不幸的是，它们往往效率很低，通常需要数小时或数天的计算。因此，这些算法提供的视觉提示(着色、阴影、透明度、反射等)仍然没有被需要可视化的应用程序常规使用，主要是因为它们在当前一代图形工作站上不被实时支持。然而，在下一代工作站中，性能和功能都有望得到提升。本研究项目开发高效、高质量的图像合成算法。该项目将攻击几个特定的领域。继续发展渲染背后的理论，使其具有坚实的物理和数学基础，继续开发基于小波的快速分层算法来求解渲染方程。研究了结合有限元和蒙特卡罗方法最佳特性的各种混合算法。最后，调查因需要处理非常大、复杂的环境而产生的各种问题。这些算法将在正在开发的实验性图像合成系统中实施，并将在各种不同的场景和环境中进行测试。

Many realistic image synthesis algorithms have been devised, but unfortunately they tend to be very inefficient, often requiring hours or days of calculation. As a result, the visual cues provided by these algorithms (shading, shadows, transparency, reflections, etc.) still have not been used routinely by applications requiring visualization, primarily because they are not supported in real-time on current generation graphics workstations. However, both increased performance and greater functionality can be expected in future generation workstation. This research project develops efficient high quality image synthesis algorithms. The project will attack several specific areas. Continuing to develop the theory behind rendering so that it has a sound basis in physics and mathematics continuing the development of fast hierarchial algorithms based on wavelets for solving the rendering equation. Investigating various hybrid algorithms that combine the best features of the finite element and Monte Carlo approaches. Finally investigating various issues that result from the need to handle very large, complex environements. The algorithms will be implemented in an experimental image synthesis system that is being developed and they will be tested on a variety of different scenes and environments.