arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2110.07600cs.AR
收藏

PointAcc:高效的点云加速器

PointAcc: Efficient Point Cloud Accelerator

Yujun Lin, Zhekai Zhang, Haotian Tang, Hanrui Wang, Song Han

摘要:

点云上的深度学习在自动驾驶和AR/VR等广泛应用中发挥着至关重要的作用。这些应用程序在边缘设备上与人实时交互,因此需要低延迟和低能耗。与将点云投影到2D空间相比,直接处理3D点云可获得更高的精度和更低的MAC数。然而,点云的极稀疏性质对硬件加速提出了挑战。例如,我们需要显式地确定非零输出并搜索非零邻居(映射操作),这在现有加速器中是不支持的。此外,需要稀疏特征的显式聚集和分散,从而导致大的数据移动开销。本文全面分析了现代点云网络在CPU/GPU/TPU上的性能瓶颈。为了解决这些挑战,我们提出了PointAcc,一种新颖的点云深度学习加速器。PointAcc将不同的映射操作映射到一个通用的基于排名的内核上,通过可配置的缓存来流式传输稀疏计算,并在时间上融合连续的密集层以减少内存占用。在4个应用程序的8个点云模型上进行评估,PointAcc比RTX 2080 Ti GPU实现了3.7倍的加速和22倍的节能。PointAcc与轻量级神经网络共同设计,在S3 DIS数据集上运行分割的准确性提高了9.1%,速度提高了100倍,可与之前的加速器Mesorasi相媲美。PointAcc为高效的点云识别铺平了道路。

英文摘要:

Deep learning on point clouds plays a vital role in a wide range of applications such as autonomous driving and AR/VR. These applications interact with people in real-time on edge devices and thus require low latency and low energy. Compared to projecting the point cloud to 2D space, directly processing the 3D point cloud yields higher accuracy and lower #MACs. However, the extremely sparse nature of point cloud poses challenges to hardware acceleration. For example, we need to explicitly determine the nonzero outputs and search for the nonzero neighbors (mapping operation), which is unsupported in existing accelerators. Furthermore, explicit gather and scatter of sparse features are required, resulting in large data movement overhead. In this paper, we comprehensively analyze the performance bottleneck of modern point cloud networks on CPU/GPU/TPU. To address the challenges, we then present PointAcc, a novel point cloud deep learning accelerator. PointAcc maps diverse mapping operations onto one versatile ranking-based kernel, streams the sparse computation with configurable caching, and temporally fuses consecutive dense layers to reduce the memory footprint. Evaluated on 8 point cloud models across 4 applications, PointAcc achieves 3.7X speedup and 22X energy savings over RTX 2080Ti GPU. Co-designed with light-weight neural networks, PointAcc rivals the prior accelerator Mesorasi by 100X speedup with 9.1% higher accuracy running segmentation on the S3DIS dataset. PointAcc paves the way for efficient point cloud recognition.

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。

↑