FuseFPS: Accelerating Farthest Point Sampling with Fusing KD-tree Construction for Point Clouds

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Han, Meng, Wang, Liang, Xiao, Limin, Zhang, Hao, Zhang, Chenhao, Xie, Xilong, Zheng, Shuai, Dong, Jin
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913285551423488
author Han, Meng
Wang, Liang
Xiao, Limin
Zhang, Hao
Zhang, Chenhao
Xie, Xilong
Zheng, Shuai
Dong, Jin
author_facet Han, Meng
Wang, Liang
Xiao, Limin
Zhang, Hao
Zhang, Chenhao
Xie, Xilong
Zheng, Shuai
Dong, Jin
contents Point cloud analytics has become a critical workload for embedded and mobile platforms across various applications. Farthest point sampling (FPS) is a fundamental and widely used kernel in point cloud processing. However, the heavy external memory access makes FPS a performance bottleneck for real-time point cloud processing. Although bucket-based farthest point sampling can significantly reduce unnecessary memory accesses during the point sampling stage, the KD-tree construction stage becomes the predominant contributor to execution time. In this paper, we present FuseFPS, an architecture and algorithm co-design for bucket-based farthest point sampling. We first propose a hardware-friendly sampling-driven KD-tree construction algorithm. The algorithm fuses the KD-tree construction stage into the point sampling stage, further reducing memory accesses. Then, we design an efficient accelerator for bucket-based point sampling. The accelerator can offload the entire bucket-based FPS kernel at a low hardware cost. Finally, we evaluate our approach on various point cloud datasets. The detailed experiments show that compared to the state-of-the-art accelerator QuickFPS, FuseFPS achieves about 4.3$\times$ and about 6.1$\times$ improvements on speed and power efficiency, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2309_05017
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle FuseFPS: Accelerating Farthest Point Sampling with Fusing KD-tree Construction for Point Clouds
Han, Meng
Wang, Liang
Xiao, Limin
Zhang, Hao
Zhang, Chenhao
Xie, Xilong
Zheng, Shuai
Dong, Jin
Hardware Architecture
Point cloud analytics has become a critical workload for embedded and mobile platforms across various applications. Farthest point sampling (FPS) is a fundamental and widely used kernel in point cloud processing. However, the heavy external memory access makes FPS a performance bottleneck for real-time point cloud processing. Although bucket-based farthest point sampling can significantly reduce unnecessary memory accesses during the point sampling stage, the KD-tree construction stage becomes the predominant contributor to execution time. In this paper, we present FuseFPS, an architecture and algorithm co-design for bucket-based farthest point sampling. We first propose a hardware-friendly sampling-driven KD-tree construction algorithm. The algorithm fuses the KD-tree construction stage into the point sampling stage, further reducing memory accesses. Then, we design an efficient accelerator for bucket-based point sampling. The accelerator can offload the entire bucket-based FPS kernel at a low hardware cost. Finally, we evaluate our approach on various point cloud datasets. The detailed experiments show that compared to the state-of-the-art accelerator QuickFPS, FuseFPS achieves about 4.3$\times$ and about 6.1$\times$ improvements on speed and power efficiency, respectively.
title FuseFPS: Accelerating Farthest Point Sampling with Fusing KD-tree Construction for Point Clouds
topic Hardware Architecture
url https://arxiv.org/abs/2309.05017