PNNPU: A 11.9 TOPS/W High-speed 3D Point Cloud-based Neural Network Processor with Block-based Point Processing for Regular DRAM Access

Abstract

An efficient and high-speed 3D point cloud-based neural network processing unit (PNNPU) is proposed using the block-based point processing. It has three key features: 1) page-based point block memory management unit (PMMU) with linked list-based page table (LLPT) for on-chip memory footprint reduction, 2) hierarchical block-wise farthest point sampling (HFPS), and block skipping ball-query (BSBQ) for fast and efficient point processing, 3) Skipping-based max-pooling prediction (SMPP) for throughput enhancement. The PNNPU is fabricated in 65nm CMOS process and evaluated on the 3D object detection (3D OD) application. As a result, it shows 84.8 fps at 266.8mW power consumption and achieving 6.6-11.9 TOPS/W energy efficiency.