mlsanityData to 7 Oct 2026

OccFeat: Self-supervised Occupancy Feature Prediction for Pretraining BEV Segmentation Networks

arXiv 2404.14027 ↗First posted 22 Apr 2024cs.CV cs.LGcited by 17

Not on the map. Why?

Abstract (arXiv)

We introduce a self-supervised pretraining method, called OccFeat, for camera-only Bird's-Eye-View (BEV) segmentation networks. With OccFeat, we pretrain a BEV network via occupancy prediction and feature distillation tasks. Occupancy prediction provides a 3D geometric understanding of the scene to the model. However, the geometry learned is class-agnostic. Hence, we add semantic information to the model in the 3D space through distillation from a self-supervised pretrained image foundation model. Models pretrained with our method exhibit improved BEV semantic segmentation performance, particularly in low-data scenarios. Moreover, empirical results affirm the efficacy of integrating feature distillation with 3D occupancy prediction in our pretraining approach. Repository: https://github.com/valeoai/Occfeat

Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 80 of its 88 references that are in the map. First posted on arXiv 22 Apr 2024.

Builds on

80 of 88 references.

References · 80

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  6. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  7. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  8. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  9. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
  10. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  11. 2020Exploring Simple Siamese Representation Learningcited by 2,853
  12. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
38 more
  1. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,360
  2. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  3. 2018Unsupervised Representation Learning by Predicting Image Rotationscited by 1,586
  4. 2022BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformerscited by 1,213
  5. 2016Colorful Image Colorizationcited by 1,113
  6. 2022Planning-oriented Autonomous Drivingcited by 1,068
  7. 2015Unsupervised Visual Representation Learning by Context Predictioncited by 1,062
  8. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,040
  9. 2018Deep Clustering for Unsupervised Learning of Visual Featurescited by 1,020
  10. 2021VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learningcited by 1,013
  11. 2023Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecturecited by 982
  12. 2021iBOT: Image BERT Pre-Training with Online Tokenizercited by 891
  13. 2022BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representationcited by 844
  14. 2019Self-Supervised Learning of Pretext-Invariant Representationscited by 806
  15. 2021BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-Viewcited by 711
  16. 2019Data-Efficient Image Recognition with Contrastive Predictive Codingcited by 672
  17. 2022BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detectioncited by 568
  18. 2022PETR: Position Embedding Transformation for Multi-View 3D Object Detectioncited by 550
  19. 2021FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detectioncited by 444
  20. 2023Tri-Perspective View for Vision-Based 3D Semantic Occupancy Predictioncited by 404
  21. 2022PETRv2: A Unified Framework for 3D Perception from Multi-Camera Imagescited by 351
  22. 2021Categorical Depth Distribution Network for Monocular 3D Object Detectioncited by 326
  23. 2022BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervisioncited by 302
  24. 2021HDMapNet: An Online HD Map Construction and Evaluation Frameworkcited by 280
  25. 2022Cross-view Transformers for real-time Map-view Semantic Segmentationcited by 264
  26. 2021Is Pseudo-Lidar needed for Monocular 3D Object detection?cited by 248
  27. 2021FIERY: Future Instance Prediction in Bird's-Eye View from Surround Monocular Camerascited by 239
  28. 2023Scene as Occupancycited by 208
  29. 2022M2BEV: Multi-Camera Joint 3D Detection and Segmentation with Unified Birds-Eye View Representationcited by 155
  30. 2022PolarFormer: Multi-camera 3D Object Detection with Polar Transformercited by 147
  31. 2022Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?cited by 128
  32. 2022Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Datacited by 126
  33. 2023Segment Any Point Cloud Sequences by Distilling Vision Foundation Modelscited by 123
  34. 2023FB-BEV: BEV Representation from Forward-Backward View Transformationscited by 98
  35. 2023Visual Point Cloud Forecasting enables Scalable Autonomous Drivingcited by 97
  36. 2022BEVPoolv2: A Cutting-edge Implementation of BEVDet Toward Deploymentcited by 79
  37. 2023UniPAD: A Universal Pre-training Paradigm for Autonomous Drivingcited by 73
  38. 2022ALSO: Automotive Lidar Self-supervision by Occupancy estimationcited by 68

50 most cited of 80 All 80 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 4; 2025: 10; 2026: 3.

17 papers.

Sort:

  1. 2024GaussTR: Foundation Model-Aligned Gaussian Transformer for Self-Supervised 3D Spatial Understandingcited by 46
  2. 2025GaussianFlowOcc: Sparse and Weakly Supervised Occupancy Estimation using Gaussian Splatting and Temporal Flowcited by 24
  3. 2025GaussRender: Learning 3D Occupancy with Gaussian Renderingcited by 17
  4. 2024GaussianPretrain: A Simple Unified 3D Gaussian Representation for Visual Pre-training in Autonomous Drivingcited by 11
  5. 2025Bridging Perspectives: Foundation Model Guided BEV Maps for 3D Object Detection and Trackingcited by 9
  6. 2024Robust Bird's Eye View Segmentation by Adapting DINOv2cited by 6
  7. 2025ShelfOcc: Native 3D Supervision beyond LiDAR for Vision-Based Occupancy Estimationcited by 6
  8. 2026OccAny: Generalized Unconstrained Urban 3D Occupancycited by 6
  9. 2025Forging Spatial Intelligence: A Roadmap of Multi-Modal Data Pre-Training for Autonomous Systemscited by 4
  10. 2024OE-BevSeg: An Object Informed and Environment Aware Multimodal Framework for Bird's-eye-view Vehicle Semantic Segmentationcited by 3
  11. 2025RendBEV: Semantic Novel View Synthesis for Self-Supervised Bird's Eye View Segmentationcited by 3
  12. 2025DIP: Unsupervised Dense In-Context Post-training of Visual Representationscited by 3
5 more
  1. 2025FishBEV: Distortion-Resilient Bird's Eye View Segmentation with Surround-View Fisheye Camerascited by 3
  2. 2025MapRF: Weakly Supervised Online HD Map Construction via NeRF-Guided Self-Trainingcited by 2
  3. 2025From Binary to Semantic: Utilizing Large-Scale Binary Occupancy Data for 3D Semantic Occupancy Predictioncited by 0
  4. 2026LiDAR-Anchored Collaborative Distillation for Robust 2D Representationscited by 0
  5. 2026MapGCLR: Geospatial Contrastive Learning of Representations for Online Vectorized HD Map Constructioncited by 0