mlsanityData to 7 Oct 2026

Searching Efficient 3D Architectures with Sparse Point-Voxel Convolution

arXiv 2007.16100 ↗First posted 31 Jul 2020cs.CVcited by 422

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 422 papers in the map (data through 7 Oct 2026). First posted on arXiv 31 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 62; 2022: 74; 2023: 108; 2024: 89; 2025: 61; 2026: 28.

422 papers.

Sort:

By area:

  1. 2022BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representationcited by 844
  2. 2023Point Transformer V3: Simpler, Faster, Strongercited by 579
  3. 2023Tri-Perspective View for Vision-Based 3D Semantic Occupancy Predictioncited by 404
  4. 2023SurroundOcc: Multi-Camera 3D Occupancy Prediction for Autonomous Drivingcited by 310
  5. 2023OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Predictioncited by 287
  6. 2021NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Videocited by 241
  7. 2023OccWorld: Learning a 3D Occupancy World Model for Autonomous Drivingcited by 230
  8. 2023CLIP2Scene: Towards Label-efficient 3D Scene Understanding by CLIPcited by 178
  9. 2022SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Viewscited by 175
  10. 2021RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentationcited by 173
  11. 2021Panoptic nuScenes: A Large-Scale Benchmark for LiDAR Panoptic Segmentation and Trackingcited by 165
  12. 2021(AF)2-S3Net: Attentive Feature Fusion with Adaptive Feature Selection for Sparse Semantic Segmentation Networkcited by 164
38 more
  1. 20222DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Cloudscited by 157
  2. 20223D Object Detection for Autonomous Driving: A Comprehensive Surveycited by 132
  3. 2021LiDAR R-CNN: An Efficient and Universal 3D Object Detectorcited by 129
  4. 2023Robo3D: Towards Robust and Reliable 3D Perception against Corruptionscited by 128
  5. 2023Rethinking Range View Representation for LiDAR Segmentationcited by 126
  6. 2023Spherical Transformer for LiDAR-based 3D Recognitioncited by 124
  7. 2023Segment Any Point Cloud Sequences by Distilling Vision Foundation Modelscited by 123
  8. 2022Delving into the Devils of Bird's-eye-view Perception: A Review, Evaluation and Recipecited by 122
  9. 2022Point-to-Voxel Knowledge Distillation for LiDAR Semantic Segmentationcited by 118
  10. 2023PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentationcited by 114
  11. 2021Panoptic-PolarNet: Proposal-free LiDAR Point Cloud Panoptic Segmentationcited by 107
  12. 2021Fast Point Transformercited by 103
  13. 2022NeRFusion: Fusing Radiance Fields for Large-Scale Scene Reconstructioncited by 99
  14. 2021QuantumNAS: Noise-Adaptive Search for Robust Quantum Circuitscited by 97
  15. 2024SceneScript: Reconstructing Scenes With An Autoregressive Structured Language Modelcited by 95
  16. 2021TPCN: Temporal Point Cloud Networks for Motion Forecastingcited by 94
  17. 20223D UX-Net: A Large Kernel Volumetric ConvNet Modernizing Hierarchical Transformer for Medical Image Segmentationcited by 94
  18. 2022LaserMix for Semi-Supervised LiDAR Semantic Segmentationcited by 92
  19. 2021Guided Point Contrastive Learning for Semi-supervised Point Cloud Semantic Segmentationcited by 87
  20. 2023Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Trainingcited by 85
  21. 2024GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Predictioncited by 83
  22. 2021VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehiclescited by 81
  23. 2023SCPNet: Semantic Scene Completion on Point Cloudcited by 80
  24. 2021EfficientLPS: Efficient LiDAR Panoptic Segmentationcited by 76
  25. 2021Transfer Learning from Synthetic to Real LiDAR Point Cloud for Semantic Segmentationcited by 76
  26. 2022MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentationcited by 76
  27. 2021M3DeTR: Multi-representation, Multi-scale, Mutual-relation 3D Object Detection with Transformerscited by 74
  28. 2022PolarMix: A General Data Augmentation Technique for LiDAR Point Cloudscited by 74
  29. 2022LidarMultiNet: Towards a Unified Multi-Task Network for LiDAR Perceptioncited by 74
  30. 2023UniPAD: A Universal Pre-training Paradigm for Autonomous Drivingcited by 73
  31. 2022TorchSparse: Efficient Point Cloud Inference Enginecited by 72
  32. 2023FlatFormer: Flattened Window Attention for Efficient Point Cloud Transformercited by 72
  33. 2023SSCBench: A Large-Scale 3D Semantic Scene Completion Benchmark for Autonomous Drivingcited by 72
  34. 2021PointMixer: MLP-Mixer for Point Cloud Understandingcited by 71
  35. 2022LargeKernel3D: Scaling up Kernels in 3D Sparse CNNscited by 71
  36. 2022Scribble-Supervised LiDAR Semantic Segmentationcited by 69
  37. 2022ALSO: Automotive Lidar Self-supervision by Occupancy estimationcited by 68
  38. 2023MSeg3D: Multi-modal 3D Semantic Segmentation for Autonomous Drivingcited by 68

50 most cited of 422 All 422 (JSON)