Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution

Abstract

Self-driving cars need to understand 3D scenes efficiently and accurately in order to drive safely. Given the limited hardware resources, existing 3D perception models are not able to recognize small instances (e.g., pedestrians, cyclists) very well due to the low-resolution voxelization and aggressive downsampling. To this end, we propose Sparse Point-Voxel Convolution (SPVConv), a lightweight 3D module that equips the vanilla Sparse Convolution with the high-resolution point-based branch. With negligible overhead, this point-based branch is able to preserve the fine details even from large outdoor scenes. To explore the spectrum of efficient 3D models, we first define a flexible architecture design space based on SPVConv, and we then present 3D Neural Architecture Search (3D-NAS) to search the optimal network architecture over this diverse design space efficiently and effectively. Experimental results validate that the resulting SPVNAS model is fast and accurate: it outperforms the state-of-the-art MinkowskiNet by 3.3%, ranking 1st on the competitive SemanticKITTI leaderboard. It also achieves 8x computation reduction and 3x measured speedup over MinkowskiNet with higher accuracy. Finally, we transfer our method to 3D object detection, and it achieves consistent improvements over the one-stage detection baseline on KITTI.

ECCV 2020. The first two authors contributed equally to this work. Project page: http://spvnas.mit.edu/

References in corpus (6)

• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• DARTS: Differentiable Architecture Search
• AMC: AutoML for Model Compression and Acceleration on Mobile Devices
• HAT: Hardware-Aware Transformers for Efficient Natural Language Processing
• SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
• APQ: Joint Search for Network Architecture, Pruning and Quantization Policy

Cited by in corpus (8)

• FPS-Net: A Convolutional Fusion Network for Large-Scale LiDAR Point Cloud Segmentation
• Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation
• Deep Learning on 3D Semantic Segmentation: A Detailed Review
• TPCN: Temporal Point Cloud Networks for Motion Forecasting
• NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video
• LiDAR R-CNN: An Efficient and Universal 3D Object Detector
• Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR-based Perception
• Predicting the success of Gradient Descent for a particular Dataset-Architecture-Initialization (DAI)