mlsanityData to 7 Oct 2026

DiffBEV: Conditional Diffusion Model for Bird's Eye View Perception

arXiv 2303.08333 ↗First posted 15 Mar 2023cs.CVcited by 30

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

BEV perception is of great importance in the field of autonomous driving, serving as the cornerstone of planning, controlling, and motion prediction. The quality of the BEV feature highly affects the performance of BEV perception. However, taking the noises in camera parameters and LiDAR scans into consideration, we usually obtain BEV representation with harmful noises. Diffusion models naturally have the ability to denoise noisy samples to the ideal data, which motivates us to utilize the diffusion model to get a better BEV representation. In this work, we propose an end-to-end framework, named DiffBEV, to exploit the potential of diffusion model to generate a more comprehensive BEV representation. To the best of our knowledge, we are the first to apply diffusion model to BEV perception. In practice, we design three types of conditions to guide the training of the diffusion model which denoises the coarse samples and refines the semantic feature in a progressive way. What's more, a cross-attention module is leveraged to fuse the context of BEV feature and the semantic content of conditional diffusion model. DiffBEV achieves a 25.9% mIoU on the nuScenes dataset, which is 6.2% higher than the best-performing existing approach. Quantitative and qualitative results on multiple benchmarks demonstrate the effectiveness of DiffBEV in BEV semantic segmentation and 3D object detection tasks. The code will be available soon.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 47 of its 53 references that are in the map. First posted on arXiv 15 Mar 2023.

Builds on

47 of 53 references.

Closest ancestors · 3

Same direction, newest first

  1. 2022DiffusionDet: Diffusion Model for Object Detection17 Nov
  2. 2022Blended Latent Diffusion6 Jun
  3. 2021DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulation6 Oct

Other references · 44

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  5. 2022Video Diffusion Modelscited by 2,280
  6. 2022Make-A-Video: Text-to-Video Generation without Text-Video Datacited by 1,725
  7. 2017Decoupled Weight Decay Regularizationcited by 1,722
  8. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
  9. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  10. 2020Center-based 3D Object Detection and Trackingcited by 1,143
  11. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,040
  12. 2022Planning with Diffusion for Flexible Behavior Synthesiscited by 1,023
32 more
  1. 2022Denoising Diffusion Restoration Modelscited by 951
  2. 2021Vector Quantized Diffusion Model for Text-to-Image Synthesiscited by 773
  3. 2021Diffusion Probabilistic Models for 3D Point Cloud Generationcited by 734
  4. 2021BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-Viewcited by 711
  5. 2021DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queriescited by 655
  6. 2019CenterNet: Keypoint Triplets for Object Detectioncited by 654
  7. 2021ILVR: Conditioning Method for Denoising Diffusion Probabilistic Modelscited by 634
  8. 2022BEVDepth: Acquisition of Reliable Depth for Multi-view 3D Object Detectioncited by 568
  9. 2022LION: Latent Point Diffusion Models for 3D Shape Generationcited by 552
  10. 2022PETR: Position Embedding Transformation for Multi-View 3D Object Detectioncited by 550
  11. 20213D Shape Generation and Completion through Point-Voxel Diffusioncited by 516
  12. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANscited by 515
  13. 2021Label-Efficient Semantic Segmentation with Diffusion Modelscited by 495
  14. 2021FCOS3D: Fully Convolutional One-Stage Monocular 3D Object Detectioncited by 444
  15. 2022BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detectioncited by 340
  16. 2021SegDiff: Image Segmentation with Diffusion Probabilistic Modelscited by 279
  17. 2022Cross-view Transformers for real-time Map-view Semantic Segmentationcited by 264
  18. 2022Diffusion models as plug-and-play priorscited by 230
  19. 2018Orthographic Feature Transform for Monocular 3D Object Detectioncited by 201
  20. 2022BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Drivingcited by 191
  21. 2021Probabilistic and Geometric Depth: Detecting Objects in Perspectivecited by 190
  22. 2021A Variational Perspective on Diffusion-Based Generative Models and Score Matchingcited by 182
  23. 2022MedSegDiff: Medical Image Segmentation with Diffusion Probabilistic Modelcited by 180
  24. 2020Predicting Semantic Map Representations from Images using Pyramid Occupancy Networkscited by 175
  25. 2019Cross-view Semantic Segmentation for Sensing Surroundingscited by 173
  26. 2022A Generalist Framework for Panoptic Segmentation of Images and Videoscited by 99
  27. 2018Monocular Semantic Occupancy Grid Mapping with Convolutional Variational Encoder-Decoder Networkscited by 84
  28. 2021Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codescited by 71
  29. 2022DiffusionInst: Diffusion Model for Instance Segmentationcited by 49
  30. 2022DiffuStereo: High Quality Human Reconstruction via Diffusion-based Stereo Using Sparse Camerascited by 43
  31. 2022GitNet: Geometric Prior-based Transformation for Birds-Eye-View Segmentationcited by 24
  32. 2022HFT: Lifting Perspective Representations via Hybrid Feature Transformationcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 4; 2024: 7; 2025: 11; 2026: 8.

30 papers.

Sort:

By area:

  1. 2025DiffAD: A Unified Diffusion Modeling Approach for Autonomous Drivingcited by 26
  2. 2024DifFUSER: Diffusion Model for Robust Multi-Sensor Fusion in 3D Object Detection and BEV Segmentationcited by 18
  3. 2025BEVDiffuser: Plug-and-Play Diffusion Model for BEV Denoising with Ground-Truth Guidancecited by 17
  4. 2025DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Drivingcited by 17
  5. 2024Improving Bird's Eye View Semantic Segmentation by Task Decompositioncited by 13
  6. 2024BLOS-BEV: Navigation Map Enhanced Lane Segmentation Network, Beyond Line of Sightcited by 12
  7. 2025MapDiffusion: Generative Diffusion for Vectorized Online HD Map Construction and Uncertainty Estimation in Autonomous Drivingcited by 12
  8. 2025Pragmatic Heterogeneous Collaborative Perception via Generative Communication Mechanismcited by 9
  9. 2024DiffCP: Ultra-Low Bit Collaborative Perception via Diffusion Modelcited by 6
  10. 2025DualDiff: Dual-branch Diffusion Model for Autonomous Driving with Semantic Fusioncited by 6
  11. 2023GUPNet++: Geometry Uncertainty Propagation Network for Monocular 3D Object Detectioncited by 5
  12. 2023Intriguing Properties of Diffusion Models: An Empirical Study of the Natural Attack Capability in Text-to-Image Generative Modelscited by 4
18 more
  1. 2025CoDiff: Conditional Diffusion Model for Collaborative 3D Object Detectioncited by 4
  2. 2023DiffRef3D: A Diffusion-based Proposal Refinement Framework for 3D Object Detectioncited by 2
  3. 2023Diffusion-Based Particle-DETR for BEV Perceptioncited by 2
  4. 2024Focus on BEV: Self-calibrated Cycle View Transformation for Monocular Birds-Eye-View Segmentationcited by 2
  5. 2024VQ-Map: Bird's-Eye-View Map Layout Estimation in Tokenized Discrete Space via Vector Quantizationcited by 2
  6. 2025RESAR-BEV: An Explainable Progressive Residual Autoregressive Approach for Camera-Radar Fusion in BEV Segmentationcited by 2
  7. 2025Diffusion-Based Generative Models for 3D Occupancy Prediction in Autonomous Drivingcited by 2
  8. 2024EffiPerception: an Efficient Framework for Various Perception Taskscited by 1
  9. 2026RESBev: Making BEV Perception More Robustcited by 1
  10. 2026Reconstruction Matters: Learning Geometry-Aligned BEV Representation through 3D Gaussian Splattingcited by 1
  11. 2026Learning Ego-Centric BEV Representations from a Perspective-Privileged View: Cross-View Supervision for Online HD Map Constructioncited by 1
  12. 2025FMOcc: TPV-Driven Flow Matching for 3D Occupancy Prediction with Selective State Space Modelcited by 0
  13. 2025DRCP: Diffusion on Reinforced Cooperative Perception for Perceiving Beyond Limitscited by 0
  14. 2026CycleBEV: Regularizing View Transformation Networks via View Cycle Consistency for Bird's-Eye-View Semantic Segmentationcited by 0
  15. 2026FlatLands: Generative Floormap Completion From a Single Egocentric Viewcited by 0
  16. 2026Generative-Model Predictive Planning for Navigation in Partially Observable Environmentscited by 0
  17. 2026BEV-Denoise: Learning Intrinsic Noise for Accurate Bird's-Eye-View Semantic Segmentationcited by 0
  18. 2026Open-Vocabulary BEV Segmentation with 3D-Aware Geometric Constraintscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Diff-2-in-1: Bridging Generation and Dense Perception with Diffusion Models
  2. 2024Diffusion Models in Low-Level Vision: A Survey
  3. 2023DDP: Diffusion Model for Dense Visual Prediction
  4. 2024Diffusion Models in 3D Vision: A Survey
  5. 2024DiffMap: Enhancing Map Segmentation with Map Prior Using Diffusion Model
  6. 2024DiffSF: Diffusion Models for Scene Flow Estimation
  7. 2022Diffusion Models in Vision: A Survey
  8. 2024Identifying and Solving Conditional Image Leakage in Image-to-Video Diffusion Model