基于GPU的光子映射并行化算法

针对串行情况下光子映射算法速度慢的问题,对光子映射算法并行化进行可行性分析,充分利用图像处理器(GPU)的统一设备计算架构(CUDA)的并行和计算能力,实现光子映射算法的并行化。同时针对算法中光子发射追踪阶段生成GPU线程数与光子数相同的方法的不足以及平均分配方法所造成的资源浪费等,提出线程之间协同工作的方法并采用动态平衡处理,使光子渲染速度提升了将近一倍。实验结果证明了多线程间协同工作及动态平衡相结合方法的有效性。     

基于自适应光子发射的渐进式光子映射

通过对渐进式光子映射算法进行扩展,提出了一种基于自适应光子发射的渐进式光子映射算法.渐进式光子映射是一个多遍的全局光照算法,通过不断发射光子并渐进更新场景各点的光能估计能使其最终能收敛到无偏差的结果.由于渐进式光子映射完全使用密度估计来计算各点的光能,因此其收敛速度受光子分布影响较大.利用渐进式光子映射算法中固有的场景统计信息以及其多遍的特点,设计了一个自适应的光子发射策略,使得发射的光子能更多的分布在对最终绘制有效的区域,提高了原算法的绘制效率.     

基于GPU的变型SIFT算子实时图像配准

针对SIFT变型算法描述向量维数过高实、时性差的问题,分别在建立高斯尺度金字塔、关键点的亚像素定位等方面进行改进与并行化。利用CUDA设备构架在GPU硬件上实现多线程,一方面避免了PCA方法造成的关键点信息流失,另一方面使得配准速度达到了工程中的实时性要求。在VS2005平台上通过C语言实现混合CUDA编程,结果表明该方法使得配准速度和点对匹配正确率都有较大提升。     

快速虚顶点光子映射的焦散绘制

目的 模拟绘制焦散效果是真实感图形绘制的重要组成部分。利用可编程GPU硬件,基于图像的光子映射方法绘制速度快,但由于使用了近似采样和计算,会产生失真。为了克服这一现象,针对理想的镜面反射、折射体,提出一种快速绘制焦散效果的方法（VOBPBT）。方法 该方法首次定义了光子路径映射图的概念,并提出通过基于虚顶点光子映射的光子束跟踪来准确构建光子路径映射图的方法;此外,方法也创新性地提出利用光子路径映射图来查找焦散三角形,创建焦散映射图的方法。结果 实验结果表明,本文绘制结果真实,可以处理多次递归反射、折射,能够绘制连贯的高频焦散效果,同时可以达到交互的绘制性能。结论 本文VOBPBT方法在可交互计算机仿真、计算机游戏、虚拟漫游等应用领域具有一定的实用价值。     

基于栅格的全局光子图重建

基于光子图的光子映射算法能产生高质量的照片级图像。对于光照复杂的 场景,光子图需要存储大量光子以提高生成图像的质量,这不仅占用大量的内存空间,而且 光照估计的时间长。论文提出基于栅格的全局光子图重建的算法,即在光子包围盒被栅格化 后,其非空栅格中一定比例的光子被用来重建新的光子图,并保证重建前后栅格内光子能量 和守恒,这使得重建前后光子图的光照估计的效果相近。通过增加特定栅格中的重建光子数 目,能有效减少由几何偏差引起的光照估计误差,增强直接聚焦（焦散）和间接聚焦光照的 绘制效果;并使用简单方法检测生成图像中少量噪声,增加少量采样即可有效减少相应的噪 声。全局光子图重建算法的计算成本低,并保持生成图像的视觉独立性。     

使用光子映射渲染参与介质

光子映射是近年发展起来的一种新的全局光照算法。本文依据光子映射对实体物体的渲染,将其扩展到对包含参与介质的场景的渲染,为此提出了一个两路的渲染算法。在第一路中,光子从光源发射,并使用光子追踪来构造体光子图;第二路从视点出发向场景中发射光线,使用光线追踪来进行渲染,其中,根据构造好的光子图,用光线步进进行
行递归的辐射估计,得出最终光强。     

GPU加速分子动力学模拟的热力学量提取*

近年来,统一计算设备架构(CUDA)的提出和图形处理器（GPU）快速提升的并行处理能力和数据传输能力,使得基于CUDA的GPU通用计算迅速成为一个研究热点。针对含有大规模分子动力学模拟的热力学量提取效率低下的问题,提出了分子动力学模拟的热力学量提取的新方法,利用CUDA设计了并行算法,实现了利用GPU加速分子动力学模拟的热力学量提取。实验结果表明,与基于CPU的算法相比, GPU可以提高速度500倍左右。     

基于光子映射的辐照度缓存算法

提出了一种全局光照计算方法,结合了两个知名的技术,光子映射和辐照度缓存.光子映射具有视点无关的优势,辐照度缓存可以快速计算间接光照,但后者是视点相关的,为了使光照缓存记录覆盖整个场景,辐照度缓存算法需要手动设置很多相机.利用这两种技术的各自优势,通过光子图来计算改进后的视点无关的辐照度缓存算法,实现了快速而准确的全局光...     

基于CUDA平台的FIR滤波算法的设计与优化

针对目前基于普通DSP的FIR算法速度低、扩展性差的缺点,提出并实现基于CUDA平台实现的FIR滤波算法。由于在CUDA中程序可以直接操作数据而无需借助于图形系统的API,使开发者能够在GPU 强大计算能力的基础上建立起一种效率更高的密集数据计算解决方案。该算法将CUDA用于FIR滤波器输入输出关系计算,采用矩阵乘法的并行运算技术,在GPU上建立并行滤波模型,并对算法进行了优化。实验结果表明,在Tesla C1060平台上,和传统的基于DSP的FIR滤波算法计算速度相比,基于CUDA平台计算FIR滤波算法时,其加速比可接近30,解决了传统基于DSP计算FIR滤波算法速度较慢、扩展性差的问题。     

基于CUDA的并行布谷鸟搜索算法设计与实现

布谷鸟搜索（cuckoo search,CS）算法是近几年发展起来的智能元启发式算法,已经被成功应用于多种优化问题中。针对CS算法在求解大数据、大规模复杂问题时,计算时间过长的问题,提出了一种基于统一计算设备架构（compute unified device architecture,CUDA）的并行布谷鸟搜索算法。该算法的并行实现采用任务并行与数据并行相结合的方式,利用图形处理器（graphic processing unit,GPU）线程块与线程分别映射布谷鸟个体与个体的每一维数据,并行实现CS算法中的鸟巢位置更新、个体适应度评估、鸟巢重建、寻找最优个体操作。整个CS算法的寻优迭代过程完全通过GPU实现,降低了算法计算过程中CPU与GPU的通信开销。对4个经典基准测试函数进行了仿真实验,结果表明,相比标准CS算法,基于CUDA架构的并行CS算法在求解收敛性一致的前提下,在求解速度上获得了高达110倍的计算加速比。     

基于CUDA的高速FFT计算*

针对快速傅里叶算法FFT在图形图像处理和科学计算领域的重要作用,提出了一种基于CUDA的高速FFT计算方法,在分析GPU硬件平台执行模式及FFT算法并行性特征的基础上,采用多线程并行的映射方法实现算法,并从存储层次优化算法。实验结果表明该算法的高效性,优化后的FFT加速比能达到CUFFT库加速比的2-6倍。     

The Beam Radiance Estimate for Volumetric Photon Mapping

We present a new method for efficiently simulating the scattering of light within participating media. Using a theoretical reformulation of volumetric photon mapping, we develop a novel photon gathering technique for participating media. Traditional volumetric photon mapping samples the in‐scattered radiance at numerous points along the length of a single ray by performing costly range queries within the photon map. Our technique replaces these multiple point‐queries with a single beam‐query, which explicitly gathers all photons along the length of an entire ray. These photons are used to estimate the accumulated in‐scattered radiance arriving from a particular direction and need to be gathered only once per ray. Our method handles both fixed and adaptive kernels, is faster than regular volumetric photon mapping, and produces images with less noise.     

Photon Differential Splatting for Rendering Caustics

We present a photon splatting technique which reduces noise and blur in the rendering of caustics. Blurring of illumination edges is an inherent problem in photon splatting, as each photon is unaware of its neighbours when being splatted. This means that the splat size is usually based on heuristics rather than knowledge of the local flux density. We use photon differentials to determine the size and shape of the splats such that we achieve adaptive anisotropic flux density estimation in photon splatting. As compared to previous work that uses photon differentials, we present the first method where no photons or beams or differentials need to be stored in a map. We also present improvements in the theory of photon differentials, which give more accurate results and a faster implementation. Our technique has good potential for GPU acceleration, and we limit the number of parameters requiring user adjustment to an overall smoothing parameter and the number of photons to be traced.     

Deep Kernel Density Estimation for Photon Mapping

Recently, deep learning-based denoising approaches have led to dramatic improvements in low sample-count Monte Carlo rendering. These approaches are aimed at path tracing, which is not ideal for simulating challenging light transport effects like caustics, where photon mapping is the method of choice. However, photon mapping requires very large numbers of traced photons to achieve high-quality reconstructions. In this paper, we develop the first deep learning-based method for particle-based rendering, and specifically focus on photon density estimation, the core of all particle-based methods. We train a novel deep neural network to predict a kernel function to aggregate photon contributions at shading points. Our network encodes individual photons into per-photon features, aggregates them in the neighborhood of a shading point to construct a photon local context vector, and infers a kernel function from the per-photon and photon local context features. This network is easy to incorporate in many previous photon mapping methods (by simply swapping the kernel density estimator) and can produce high-quality reconstructions of complex global illumination effects like caustics with an order of magnitude fewer photons compared to previous photon mapping methods. Our approach largely reduces the required number of photons, significantly advancing the computational efficiency in photon mapping.     

Distributed Out‐of‐Core Stochastic Progressive Photon Mapping

At present, stochastic progressive photon mapping (SPPM) is one of the most comprehensive methods for a consistent global illumination computation. Even though the number of photons is unlimited due to their progressive nature, the scene size is still bound by the available main memory. In this paper, we present the first consistent out‐of‐core SPPM algorithm. In order to cope with large scenes, we automatically subdivide the geometry and parallelly trace photons and eye rays in a portal‐based system, distributed across multiple machines in a commodity cluster. Moreover, modifications of the original SPPM method are introduced that keep both the utilization of tracer machines high and the network traffic low. Therefore, compared to a portal‐based single machine setup, our distributed approach achieves a significant speedup. We compare a GPU‐based with a CPU‐based implementation and demonstrate our system in multiple large test scenes of up to 90 million triangles.     

Grouped photon mapping

This paper proposes a novel architecture called Grouped Photon Mapping, which combines standard photon mapping with the light-beam concept to improve the nearest-neighbor density estimation method. Based on spatial coherence, we cluster all of photons, which are deposited in the photon map, into different beam-like groups. Each group of photons is individually stored in an isolated photon map. By the distribution of the photons in each photon map, we construct a polygonal boundary to represent a beam-like illuminated area. These boundaries offer a more accurate and flexible sampling area to filter neighbor photons around the query point. In addition, by a level of detail technique, we can control the photon-count in each group to obtain a balance between biases and noise. The results of our experiments prove that our method can successfully reduce bias errors and light leakage. Especially, for complicated caustic effects through a gemstone-like object, we can render a smoother result than standard photon mapping.