ICARUS:一种用于神经辐射场绘制的专用体系结构
ICARUS: A Specialized Architecture for Neural Radiance Fields Rendering
摘要:
神经辐射场(NeRF)在渲染应用中的实际部署面临着几个挑战,其中最关键的一个是即使在高端图形处理单元(GPU)上的低渲染速度。在本文中,我们提出了ICARUS,一个专门的加速器架构为NeRF渲染量身定制。与使用NeRF通用计算和存储器架构的GPU不同,ICARUS使用专用全光核心(PLCore)执行完整的NeRF流水线,PLCore由位置编码单元(PEU)、多层感知器(MLP)引擎和体渲染单元(VRU)组成。PLCore接收位置\&方向并渲染相应的像素颜色,而没有任何中间数据离开芯片进行临时存储和交换,这可能是耗时和功耗的。为了实现NeRF最昂贵的组件,即,在MLP的基础上,我们将全连接运算转化为近似的可重构多常数乘法(MCMs),不同乘法之间共享公共子表达式,以提高计算效率。我们建立了一个原型ICARUS使用Synopsys公司的HAPS-80 S104,现场可编程门阵列(FPGA)为基础的大规模集成电路和系统设计的原型系统。我们评估了使用40纳米LP CMOS技术的PLCore的功率性能面积(PPA)。工作在400 MHz时,单个PLCore占用16.5 $mm^2$,消耗282.8 mW,转换为0.105 uJ/样本。将结果与GPU和张量处理单元(TPU)实现的结果进行比较。
英文摘要:
The practical deployment of Neural Radiance Fields (NeRF) in rendering applications faces several challenges, with the most critical one being low rendering speed on even high-end graphic processing units (GPUs). In this paper, we present ICARUS, a specialized accelerator architecture tailored for NeRF rendering. Unlike GPUs using general purpose computing and memory architectures for NeRF, ICARUS executes the complete NeRF pipeline using dedicated plenoptic cores (PLCore) consisting of a positional encoding unit (PEU), a multi-layer perceptron (MLP) engine, and a volume rendering unit (VRU). A PLCore takes in positions \& directions and renders the corresponding pixel colors without any intermediate data going off-chip for temporary storage and exchange, which can be time and power consuming. To implement the most expensive component of NeRF, i.e., the MLP, we transform the fully connected operations to approximated reconfigurable multiple constant multiplications (MCMs), where common subexpressions are shared across different multiplications to improve the computation efficiency. We build a prototype ICARUS using Synopsys HAPS-80 S104, a field programmable gate array (FPGA)-based prototyping system for large-scale integrated circuits and systems design. We evaluate the power-performance-area (PPA) of a PLCore using 40nm LP CMOS technology. Working at 400 MHz, a single PLCore occupies 16.5 $mm^2$ and consumes 282.8 mW, translating to 0.105 uJ/sample. The results are compared with those of GPU and tensor processing unit (TPU) implementations.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。