PointAcc: Efficient Point Cloud Accelerator

Abstract

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.

arXiv commentAccepted by Mircro 2021

References in corpus (2)

• FINN: A Framework for Fast, Scalable Binarized Neural Network Inference
• Tigris: Architecture and Algorithms for 3D Perception in Point Clouds

Cited by in corpus (6)

• A Composable Dynamic Sparse Dataflow Architecture for Efficient Event-based Vision Processing on FPGA
• GNNBuilder: An Automated Framework for Generic Graph Neural Network Accelerator Generation, Simulation, and Optimization
• K-D Bonsai: ISA-Extensions to Compress K-D Trees for Autonomous Driving Tasks
• FuseFPS: Accelerating Farthest Point Sampling with Fusing KD-tree Construction for Point Clouds
• HgPCN: A Heterogeneous Architecture for E2E Embedded Point Cloud Inference
• LEGO: Spatial Accelerator Generation and Optimization for Tensor Applications