Research map / Video / Video generation
DriveDreamer: Towards Real-world-driven World Models for Autonomous Driving
Xiaofeng Wang, Zheng Zhu, Guan Huang, Chen Xinze, Jiagang Zhu, Jiwen Lu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add first-stage training (structured traffic conditions) to Auto-DM (ablation) | DriveDreamer · nuScenes val generation | FID / FVD: FID 15.9 -> 15.3; FVD 363.3 -> 349.6 | Improved |
| add second-stage ActionFormer training (ablation) | DriveDreamer · nuScenes val generation | FID / FVD: FID 15.3 -> 14.9; FVD 349.6 -> 340.8 | Improved |
| use diffusion backbone instead of DriveGAN, even without first-stage training | DriveDreamer · nuScenes val generation | FID / FVD: FID 27.8 -> 15.9; FVD 390.8 -> 363.3 | Improved |
| add 4K synthetic images to FCOS3D training | FCOS3D · nuScenes 3D detection | mAP: 30.2 -> 30.9 | Improved |
| add 4K synthetic images to BEVFusion training | BEVFusion · nuScenes 3D detection | mAP: 32.8 -> 35.8 | Improved |
| add synthetic data to FCOS3D training | FCOS3D · nuScenes 3D detection | NDS: 38.1 -> 38.3 | No clear effect |
| add visual features from Auto-DM to action-based planning | DriveDreamer · nuScenes open-loop planning | L2 Avg. (m): VAD 0.37 -> 0.29 L2; collision 0.14 vs 0.15 | Mixed |
Later papers that compared with this method
10 later accepted papers report DriveDreamer as a baseline in their results tables; 10 of them report a better number on most of the results they share with it.
- RAYNOVA: Scale-Temporal Autoregressive World Modeling in Ray Space (CVPR 2026 · better on 2 of 2 shared results)
- Rethinking Driving World Model as Synthetic Data Generator for Perception Tasks (ICLR 2026 · better on 1 of 1 shared result)
- DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation (AAAI 2025 · better on 6 of 6 shared results)
- ConsistentCity: Semantic Flow-guided Occupancy DiT for Temporally Consistent Driving Scene Synthesis (ICCV 2025 · better on 2 of 2 shared results)
- NoiseController: Towards Consistent Multi-view Video Generation via Noise Decomposition and Collaboration (ICCV 2025 · better on 2 of 2 shared results)
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · better on 2 of 2 shared results)
- Generalized Predictive Model for Autonomous Driving (CVPR 2024 · better on 2 of 2 shared results)
- Panacea: Panoramic and Controllable Video Generation for Autonomous Driving (CVPR 2024 · better on 2 of 2 shared results)
2 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DriveGAN — DriveGAN: Towards a Controllable High-Quality Neural Simulation (CVPR 2021) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation (CVPR 2025 · similarity 0.92)
- DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation (AAAI 2025 · similarity 0.92)
- ReSim: Reliable World Simulation for Autonomous Driving (NeurIPS 2025 · similarity 0.90)
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · similarity 0.90)
- DrivingDojo Dataset: Advancing Interactive and Knowledge-Enriched Driving World Model (NeurIPS 2024 · similarity 0.90)
- Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability (NeurIPS 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
