PC2IM: An Efficient In-Memory Computing Accelerator for 3D Point Cloud

TL;DR

This paper introduces PC2IM, a novel SRAM-based in-memory computing accelerator designed to significantly improve the speed and energy efficiency of 3D point cloud neural networks by reducing memory access and computation latency.

Contribution

The paper presents a new SRAM-CIM architecture with specialized modules for data preprocessing, distance computation, and feature extraction tailored for 3D point cloud processing.

Findings

1.5x speedup over state-of-the-art accelerators

2.7x energy efficiency improvement compared to existing solutions

3.5x speedup and 1518.9x energy efficiency over GPU implementations

Abstract

3D point cloud neural networks have significantly enhanced the perceptual capabilities of resource-limited mobile intelligent systems. However, despite the transformative impact, the point cloud algorithm suffers from substantial memory access during data preprocessing and imposes a burdensome workload on feature computing, resulting in high energy consumption and latency. In this paper, an efficient SRAM-based computing-in-memory (SRAM-CIM) accelerator (PC2IM), is proposed to alleviate memory access bottlenecks in point-based 3D point cloud networks. A data preprocessing module driven by the customized CIM engines is proposed and incorporated into a memory-efficient data flow. Specifically, an approximate distance SRAM-CIM (APD-CIM) is introduced to eliminate the repetitive on-chip memory access for point clouds that are spatially partitioned by the median and reduce the volume of temporary distance data. Building on the APD-CIM, a two-level Ping-Pong-MAX Content Addressable Memory (Ping-Pong-MAX CAM) is introduced to adaptively update temporary distances and perform in-situ search for the maximum, further reducing memory access. Additionally, an efficient CIM-based feature computing engine, named split-concatenate SRAM-CIM, is presented to minimize computation latency in multi-layer perceptron with high-precision input, while maintaining high area and energy efficiency. Experiment results show that the proposed PC2IM demonstrates 1.5x speedup and 2.7x enhanced energy efficiency compared to state-of-the-art point cloud accelerator. Moreover, PC2IM achieves 3.5x speedup and 1518.9x enhanced energy efficiency compared to GPU implementations.