mlsanityData to 14 Oct 2026

ProposalContrast: Unsupervised Pre-training for LiDAR-based 3D Object Detection

arXiv 2207.12654 ↗First posted 26 Jul 2022cs.CVcited by 43

Not on the map. Why?

Abstract (arXiv)

Existing approaches for unsupervised point cloud pre-training are constrained to either scene-level or point/voxel-level instance discrimination. Scene-level methods tend to lose local details that are crucial for recognizing the road objects, while point/voxel-level methods inherently suffer from limited receptive field that is incapable of perceiving large objects or context environments. Considering region-level representations are more suitable for 3D object detection, we devise a new unsupervised point cloud pre-training framework, called ProposalContrast, that learns robust 3D representations by contrasting region proposals. Specifically, with an exhaustive set of region proposals sampled from each point cloud, geometric point relations within each proposal are modeled for creating expressive proposal representations. To better accommodate 3D detection properties, ProposalContrast optimizes with both inter-cluster and inter-proposal separation, i.e., sharpening the discriminativeness of proposal representations across semantic classes and object instances. The generalizability and transferability of ProposalContrast are verified on various 3D detectors (i.e., PV-RCNN, CenterPoint, PointPillars and PointRCNN) and datasets (i.e., KITTI, Waymo and ONCE).

Key facts. Cited by 43 papers in the map (data through 14 Oct 2026). Builds on 37 of its 47 references that are in the map. First posted on arXiv 26 Jul 2022.

Builds on

37 of 47 references.

References · 37

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  6. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  8. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  9. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  10. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  11. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  12. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,451
25 more
  1. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  2. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  3. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,715
  4. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,273
  5. 2020Center-based 3D Object Detection and Trackingcited by 1,144
  6. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 941
  7. 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 891
  8. 2019Self-supervised Visual Feature Learning with Deep Neural Networks: A Surveycited by 657
  9. 2020PointContrast: Unsupervised Pre-training for 3D Point Cloud Understandingcited by 504
  10. 2014Semi-Supervised Learning with Deep Generative Modelscited by 496
  11. 2019PointPainting: Sequential Fusion for 3D Object Detectioncited by 449
  12. 20203DSSD: Point-based 3D Single Stage Object Detectorcited by 440
  13. 2019From Points to Parts: 3D Object Detection from Point Cloud with Part-aware and Part-aggregation Networkcited by 368
  14. 2019S4L: Self-Supervised Semi-Supervised Learningcited by 335
  15. 2021Exploring Cross-Image Pixel Contrast for Semantic Segmentationcited by 242
  16. 2021One Million Scenes for Autonomous Driving: ONCE Datasetcited by 218
  17. 2021Self-Supervised Pretraining of 3D Features on any Point-Cloudcited by 217
  18. 2021Spatio-temporal Self-Supervised Representation Learning for 3D Point Cloudscited by 180
  19. 2019Joint-task Self-supervised Learning for Temporal Correspondencecited by 80
  20. 2020Learning Representations by Predicting Bags of Visual Wordscited by 73
  21. 2020Weakly Supervised 3D Object Detection from Lidar Point Cloudcited by 67
  22. 2020LiDAR-based Online 3D Video Object Detection with Graph-based Message Passing and Spatiotemporal Transformer Attentioncited by 65
  23. 2021MultiSiam: Self-supervised Multi-instance Siamese Representation Learning for Autonomous Drivingcited by 41
  24. 2021RandomRooms: Unsupervised Pre-training from Synthetic Shapes and Randomized Layouts for 3D Object Detectioncited by 37
  25. 2021Self-Supervised Visual Representations Learning by Contrastive Mask Predictioncited by 31

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 2; 2023: 16; 2024: 18; 2025: 6; 2026: 1.

43 papers.

Sort:

By area:

  1. 2023UniPAD: A Universal Pre-training Paradigm for Autonomous Drivingcited by 73
  2. 2022GD-MAE: Generative Decoder for MAE Pre-training on LiDAR Point Cloudscited by 57
  3. 2023MV-JAR: Masked Voxel Jigsaw and Reconstruction for LiDAR-Based Self-Supervised Pre-Trainingcited by 26
  4. 2023Grid-Centric Traffic Scenario Perception for Autonomous Driving: A Comprehensive Reviewcited by 25
  5. 2024Forging Vision Foundation Models for Autonomous Driving: Challenges, Methodologies, and Opportunitiescited by 24
  6. 2023Spatiotemporal Self-supervised Learning for Point Clouds in the Wildcited by 23
  7. 2023Clustering based Point Cloud Representation Learning for 3D Analysiscited by 23
  8. 2023Boosting Video Object Segmentation via Space-time Correspondence Learningcited by 21
  9. 2023Self-supervised Learning for Pre-Training 3D Point Clouds: A Surveycited by 21
  10. 2023AD-PT: Autonomous Driving Pre-Training with Large-scale Point Cloud Datasetcited by 21
  11. 2023Weakly Supervised Monocular 3D Object Detection using Multi-View Projection and Direction Consistencycited by 18
  12. 20253UR-LLM: An End-to-End Multimodal Large Language Model for 3D Scene Understandingcited by 16
31 more
  1. 2023CALICO: Self-Supervised Camera-LiDAR Contrastive Pre-training for BEV Perceptioncited by 14
  2. 2023Self-supervised Pre-training with Masked Shape Prediction for 3D Scene Understandingcited by 13
  3. 2024MixSup: Mixed-grained Supervision for Label-efficient LiDAR-based 3D Object Detectioncited by 9
  4. 2024Shelf-Supervised Cross-Modal Pre-Training for 3D Object Detectioncited by 9
  5. 2024Harnessing Uncertainty-aware Bounding Boxes for Unsupervised 3D Object Detectioncited by 9
  6. 2024Shape2Scene: 3D Scene Representation Learning Through Pre-training on Shape Datacited by 8
  7. 2023PRED: Pre-training via Semantic Rendering on LiDAR Point Cloudscited by 7
  8. 2024Self-supervised Learning of LiDAR 3D Point Clouds via 2D-3D Neural Calibrationcited by 7
  9. 2023SPOT: Scalable 3D Pre-training via Occupancy Prediction for Learning Transferable 3D Representationscited by 5
  10. 2023Pre-Training LiDAR-Based 3D Object Detectors Through Colorizationcited by 5
  11. 2024Fine-grained Image-to-LiDAR Contrastive Distillation with Visual Foundation Modelscited by 5
  12. 2024Sense Less, Generate More: Pre-training LiDAR Perception with Masked Autoencoders for Ultra-Efficient 3D Sensingcited by 5
  13. 2024Cross-Modal Self-Supervised Learning with Effective Contrastive Units for LiDAR Point Cloudscited by 4
  14. 2024Learning 3D Perception from Others' Predictionscited by 4
  15. 2025Forging Spatial Intelligence: A Roadmap of Multi-Modal Data Pre-Training for Autonomous Systemscited by 4
  16. 2023PatchContrast: Self-Supervised Pre-training for 3D Object Detectioncited by 3
  17. 2023Fuzzy-NMS: Improving 3D Object Detection with Fuzzy Classification in NMScited by 3
  18. 2024MsSVT++: Mixed-scale Sparse Voxel Transformer with Center Voting for 3D Object Detectioncited by 3
  19. 2024Vision-Language Guidance for LiDAR-based Unsupervised 3D Object Detectioncited by 3
  20. 2024Is Contrastive Distillation Enough for Learning Comprehensive 3D Representations?cited by 3
  21. 2025CleverDistiller: Simple and Spatially Consistent Cross-modal Distillationcited by 3
  22. 2024Cohere3D: Exploiting Temporal Coherence for Unsupervised Representation Learning of Vision-based Autonomous Drivingcited by 2
  23. 2025Weakly and Self-Supervised Class-Agnostic Motion Prediction for Autonomous Drivingcited by 2
  24. 2024Boosting 3D Object Detection with Semantic-Aware Multi-Branch Frameworkcited by 1
  25. 2024TraIL-Det: Transformation-Invariant Local Feature Networks for 3D LiDAR Object Detection with Unsupervised Pre-Trainingcited by 1
  26. 2024DyConfidMatch: Dynamic Thresholding and Re-sampling for 3D Semi-supervised Learningcited by 1
  27. 2022Effective Utilisation of Multiple Open-Source Datasets to Improve Generalisation Performance of Point Cloud Segmentation Modelscited by 0
  28. 2024Finetuning Pre-trained Model with Limited Data for LiDAR-based 3D Object Detection by Bridging Domain Gapscited by 0
  29. 2025Intrinsic-feature-guided 3D Object Detectioncited by 0
  30. 2025TransBridge: Boost 3D Object Detection by Scene-Level Completion with Transformer Decodercited by 0
  31. 2026LiDAR-Anchored Collaborative Distillation for Robust 2D Representationscited by 0