mlsanityData to 7 Oct 2026

Submanifold Sparse Convolutional Networks

arXiv 1706.01307 ↗First posted 5 Jun 2017cs.NE cs.CVcited by 223

Not on the map. Why?

Abstract (arXiv)

Convolutional network are the de-facto standard for analysing spatio-temporal data such as images, videos, 3D shapes, etc. Whilst some of this data is naturally dense (for instance, photos), many other data sources are inherently sparse. Examples include pen-strokes forming on a piece of paper, or (colored) 3D point clouds that were obtained using a LiDAR scanner or RGB-D camera. Standard "dense" implementations of convolutional networks are very inefficient when applied on such sparse data. We introduce a sparse convolutional operation tailored to processing sparse data that differs from prior work on sparse convolutional networks in that it operates strictly on submanifolds, rather than "dilating" the observation with every layer in the network. Our empirical analysis of the resulting submanifold sparse convolutional networks shows that they perform on par with state-of-the-art methods whilst requiring substantially less computation.

Key facts. Cited by 223 papers in the map (data through 7 Oct 2026). First posted on arXiv 5 Jun 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 38; 2022: 53; 2023: 54; 2024: 38; 2025: 23; 2026: 17.

223 papers.

Sort:

By area:

  1. 2023ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoderscited by 703
  2. 2022Kubric: A scalable dataset generatorcited by 373
  3. 2022Mask3D: Mask Transformer for 3D Semantic Instance Segmentationcited by 299
  4. 2021PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detectioncited by 232
  5. 2024Mini-Splatting: Representing Scenes with a Constrained Number of Gaussianscited by 208
  6. 2022Unifying Voxel-based Representation with Transformer for 3D Object Detectioncited by 202
  7. 2021ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrationscited by 193
  8. 2021Embracing Single Stride 3D Object Detector with Sparse Transformercited by 193
  9. 2022PLA: Language-Driven Open-Vocabulary 3D Scene Understandingcited by 176
  10. 2022Stratified Transformer for 3D Point Cloud Segmentationcited by 166
  11. 2021Deep Parametric Continuous Convolutional Neural Networkscited by 161
  12. 2022GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Partscited by 135
38 more
  1. 2022Not All Points Are Equal: Learning Highly Efficient Point-based Detectors for 3D LiDAR Point Cloudscited by 131
  2. 2023Spherical Transformer for LiDAR-based 3D Recognitioncited by 124
  3. 20214DComplete: Non-Rigid Motion Estimation Beyond the Observable Surfacecited by 123
  4. 2022Delving into the Devils of Bird's-eye-view Perception: A Review, Evaluation and Recipecited by 122
  5. 2023DSVT: Dynamic Sparse Voxel Transformer with Rotated Setscited by 122
  6. 2021NeSF: Neural Semantic Fields for Generalizable Semantic Segmentation of 3D Scenescited by 119
  7. 2021RSN: Range Sparse Net for Efficient, Accurate LiDAR 3D Object Detectioncited by 115
  8. 2023FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimationcited by 114
  9. 2023PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentationcited by 114
  10. 2022ConvMAE: Masked Convolution Meets Masked Autoencoderscited by 105
  11. 2023Neural Kernel Surface Reconstructioncited by 102
  12. 2022PillarNet: Real-Time and High-Performance Pillar-based 3D Object Detectioncited by 99
  13. 2022SWFormer: Sparse Window Transformer for 3D Object Detection in Point Cloudscited by 92
  14. 2023Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modelingcited by 92
  15. 2021AFDetV2: Rethinking the Necessity of the Second Stage for Object Detection from Point Cloudscited by 88
  16. 2022Surface Representation for Point Cloudscited by 84
  17. 2021Spatially-Adaptive Image Restoration using Distortion-Guided Networkscited by 77
  18. 2022MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentationcited by 76
  19. 2022LidarMultiNet: Towards a Unified Multi-Task Network for LiDAR Perceptioncited by 74
  20. 2021PointMixer: MLP-Mixer for Point Cloud Understandingcited by 71
  21. 2021QueryDet: Cascaded Sparse Query for Accelerating High-Resolution Small Object Detectioncited by 70
  22. 2021Sparse Tensor-based Multiscale Representation for Point Cloud Geometry Compressioncited by 69
  23. 2022MSMDFusion: Fusing LiDAR and Camera at Multiple Scales with Multi-Depth Seeds for 3D Object Detectioncited by 62
  24. 2022CAGroup3D: Class-Aware Grouping for 3D Object Detection on Point Cloudscited by 58
  25. 2022Contrastive Boundary Learning for Point Cloud Segmentationcited by 56
  26. 2024SCube: Instant Large-Scale Scene Reconstruction using VoxSplatscited by 56
  27. 2022CoSMix: Compositional Semantic Mix for Domain Adaptation in 3D LiDAR Segmentationcited by 54
  28. 2021CaTGrasp: Learning Category-Level Task-Relevant Grasping in Clutter from Simulationcited by 49
  29. 2022VISTA: Boosting 3D Object Detection via Dual Cross-VIew SpaTial Attentioncited by 48
  30. 2023SMURF: Spatial Multi-Representation Fusion for 3D Object Detection with 4D Imaging Radarcited by 48
  31. 2023Understanding Imbalanced Semantic Segmentation Through Neural Collapsecited by 46
  32. 2021Keypoints-Based Deep Feature Fusion for Cooperative Vehicle Detection of Autonomous Drivingcited by 45
  33. 2023HEDNet: A Hierarchical Encoder-Decoder Network for 3D Object Detection in Point Cloudscited by 45
  34. 2021Multimodal End-to-End Sparse Model for Emotion Recognitioncited by 40
  35. 2023PartManip: Learning Cross-Category Generalizable Part Manipulation Policy from Point Cloud Observationscited by 40
  36. 2021To the Point: Efficient 3D Object Detection in the Range Image with Graph Convolution Kernelscited by 39
  37. 2023FastPillars: A Deployment-friendly Pillar-based 3D Detectorcited by 39
  38. 2023Lowis3D: Language-Driven Open-World Instance-Level 3D Scene Understandingcited by 38

50 most cited of 223 All 223 (JSON)