Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution
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.
Code (6)
Tasks
3D Object Detection3D Semantic SegmentationLIDAR Semantic SegmentationNeural Architecture Searchobject-detectionObject DetectionRobust 3D Semantic SegmentationSelf-Driving CarsMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
PTC-Net: Point-Wise Transformer with Sparse Convolution Network for Place Recognition
In the point-cloud-based place recognition area, the existing hybrid architectures combining both convolutional networks and transformers have shown promising performance. They mainly apply the voxel-wise transformer aft…
Point Cloud RetrievalRetrievalInterpolation-Aware Padding for 3D Sparse Convolutional Neural Networks
Sparse voxel-based 3D convolutional neural networks (CNNs) are widely used for various 3D vision tasks. Sparse voxel-based 3D CNNs create sparse non-empty voxels from the 3D input and perform 3D convolution operations on…
SegmentationSemantic SegmentationPCSCNet: Fast 3D Semantic Segmentation of LiDAR Point Cloud for Autonomous Car using Point Convolution and Sparse Convolution Network
The autonomous car must recognize the driving environment quickly for safe driving. As the Light Detection And Range (LiDAR) sensor is widely used in the autonomous car, fast semantic segmentation of LiDAR point cloud, w…
3D Semantic SegmentationSegmentationSemantic SegmentationMASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation
We propose a new approach for 3D instance segmentation based on sparse convolution and point affinity prediction, which indicates the likelihood of two points belonging to the same instance. The proposed network, built u…
3D Instance SegmentationClusteringInstance SegmentationSegmentation+1PCAC-GAN: A Sparse-Tensor-Based Generative Adversarial Network for 3D Point Cloud Attribute Compression
Learning-based methods have proven successful in compressing geometric information for point clouds. For attribute compression, however, they still lag behind non-learning-based methods such as the MPEG G-PCC standard. T…
AttributeComputational EfficiencyGenerative Adversarial Network