mlsanityData to 7 Oct 2026

Spatial Pruned Sparse Convolution for Efficient 3D Object Detection

arXiv 2209.14201 ↗First posted 28 Sep 2022cs.CVcited by 26

Not on the map. Why?

Abstract (arXiv)

3D scenes are dominated by a large number of background points, which is redundant for the detection task that mainly needs to focus on foreground objects. In this paper, we analyze major components of existing sparse 3D CNNs and find that 3D CNNs ignore the redundancy of data and further amplify it in the down-sampling process, which brings a huge amount of extra and unnecessary computational overhead. Inspired by this, we propose a new convolution operator named spatial pruned sparse convolution (SPS-Conv), which includes two variants, spatial pruned submanifold sparse convolution (SPSS-Conv) and spatial pruned regular sparse convolution (SPRS-Conv), both of which are based on the idea of dynamically determining crucial areas for redundancy reduction. We validate that the magnitude can serve as important cues to determine crucial areas which get rid of the extra computations of learning-based methods. The proposed modules can easily be incorporated into existing sparse 3D CNNs without extra architectural modifications. Extensive experiments on the KITTI, Waymo and nuScenes datasets demonstrate that our method can achieve more than 50% reduction in GFLOPs without compromising the performance.

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Builds on 36 of its 42 references that are in the map. First posted on arXiv 28 Sep 2022.

Builds on

36 of 42 references.

References · 36

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  3. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,436
  4. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,707
  5. 2017Deformable Convolutional Networkscited by 1,481
  6. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  7. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  8. 2020Center-based 3D Object Detection and Trackingcited by 1,143
  9. 2016Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfercited by 1,059
  10. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 939
  11. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 890
  12. 20203DSSD: Point-based 3D Single Stage Object Detectorcited by 440
24 more
  1. 2020Voxel R-CNN: Towards High Performance Voxel-based 3D Object Detectioncited by 409
  2. 2019From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Networkcited by 366
  3. 2019Class-balanced Grouping and Sampling for Point Cloud 3D Object Detectioncited by 350
  4. 2021Voxel Transformer for 3D Object Detectioncited by 247
  5. 2021PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detectioncited by 232
  6. 2017Submanifold Sparse Convolutional Networkscited by 223
  7. 2021Multimodal Virtual Point 3D Detectioncited by 166
  8. 2022Focal Sparse Convolutional Networks for 3D Object Detectioncited by 142
  9. 2020SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registrationcited by 141
  10. 2021Focal and Global Knowledge Distillation for Detectorscited by 138
  11. 2020CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloudcited by 107
  12. 2021FusionPainting: Multimodal Fusion with Adaptive Attention for 3D Object Detectioncited by 92
  13. 2019Dynamic Convolutions: Exploiting Spatial Sparsity for Faster Inferencecited by 91
  14. 2019SESS: Self-Ensembling Semi-Supervised 3D Object Detectioncited by 89
  15. 2019Object as Hotspots: An Anchor-Free 3D Object Detection Approach via Firing of Hotspotscited by 84
  16. 2020HVNet: Hybrid Voxel Network for LiDAR Based 3D Object Detectioncited by 82
  17. 2021Pyramid R-CNN: Towards Better Performance and Adaptability for 3D Object Detectioncited by 79
  18. 2017More is Less: A More Complicated Network with Less Inference Complexitycited by 66
  19. 2018SBNet: Sparse Blocks Network for Fast Inferencecited by 53
  20. 2020Spatially Adaptive Inference with Stochastic Feature Sampling and Interpolationcited by 51
  21. 2014Spatially-sparse convolutional neural networkscited by 33
  22. 2020Associate-3Ddet: Perceptual-to-Conceptual Association for 3D Point Cloud Object Detectioncited by 31
  23. 2020SegVoxelNet: Exploring Semantic Context and Depth-aware Features for 3D Vehicle Detection from Point Cloudcited by 16
  24. 2020Deformable PV-RCNN: Improving 3D Object Detection with Learned Deformationscited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 11; 2024: 10; 2025: 1; 2026: 4.

26 papers.

Sort:

By area:

  1. 2023VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Trackingcited by 221
  2. 2023Spherical Transformer for LiDAR-based 3D Recognitioncited by 124
  3. 2023PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentationcited by 114
  4. 2023SparseViT: Revisiting Activation Sparsity for Efficient High-Resolution Vision Transformercited by 60
  5. 2023Mask-Attention-Free Transformer for 3D Instance Segmentationcited by 45
  6. 2024OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentationcited by 43
  7. 2023FastPillars: A Deployment-friendly Pillar-based 3D Detectorcited by 39
  8. 2024LION: Linear Group RNN for 3D Object Detection in Point Cloudscited by 38
  9. 2024Unified Language-driven Zero-shot Domain Adaptationcited by 30
  10. 2023ByteTrackV2: 2D and 3D Multi-Object Tracking by Associating Every Detection Boxcited by 20
  11. 2023Boosting Few-shot 3D Point Cloud Segmentation via Query-Guided Enhancementcited by 19
  12. 2023Ada3D : Exploiting the Spatial Redundancy with Adaptive Inference for Efficient 3D Object Detectioncited by 6
14 more
  1. 2024STAL3D: Unsupervised Domain Adaptation for 3D Object Detection via Collaborating Self-Training and Adversarial Learningcited by 6
  2. 20233D Small Object Detection with Dynamic Spatial Pruningcited by 4
  3. 2024Sparse Refinement for Efficient High-Resolution Semantic Segmentationcited by 3
  4. 2024Selectively Dilated Convolution for Accuracy-Preserving Sparse Pillar-based Embedded 3D Object Detectioncited by 3
  5. 2024VALO: A Versatile Anytime Framework for LiDAR-based Object Detection Deep Neural Networkscited by 3
  6. 2024DM3D: Distortion-Minimized Weight Pruning for Lossless 3D Object Detectioncited by 2
  7. 2023Generalized Robot 3D Vision-Language Model with Fast Rendering and Pre-Training Vision-Language Alignmentcited by 1
  8. 2024Efficient 3D Perception on Multi-Sweep Point Cloud with Gumbel Spatial Pruningcited by 1
  9. 2024AlterMOMA: Fusion Redundancy Pruning for Camera-LiDAR Fusion Models with Alternative Modality Maskingcited by 0
  10. 2025GTAD: Global Temporal Aggregation Denoising Learning for 3D Semantic Occupancy Predictioncited by 0
  11. 2026SUG-Occ: Explicit Semantics and Uncertainty Guided Sparse Learning for Efficient 3D Occupancy Predictioncited by 0
  12. 20262K Retrofit: Entropy-Guided Efficient Sparse Refinement for High-Resolution 3D Geometry Predictioncited by 0
  13. 2026On Exploring Input Resolution Scaling For Anytime LiDAR Object Detectioncited by 0
  14. 2026ROI-Gated SAHI: Content-Adaptive Slicing-Based Inference for Efficient Object Detectioncited by 0