Research map / Video / Video generation
InstaDrive: Instance-Aware Driving World Models for Realistic and Consistent Video Generation
Zhuoran Yang, Xi Guo, Chenjing Ding, Chiyu Wang, Wei Wu, Yanyong Zhang
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 Instance Flow Guider (ablation) | InstaDrive · nuScenes val (T+I)2V | FVD: 58.86 -> 38.06 | Improved |
| add Instance Flow Guider (ablation) | InstaDrive · nuScenes val (T+I)2V | IDS: 1189 -> 532 | Improved |
| add Instance Flow Guider (ablation) | InstaDrive · nuScenes val (T+I)2V | FID: 6.28 -> 3.96 | Improved |
| add Instance Flow Guider (ablation) | InstaDrive · nuScenes val (T+I)2V | NDS: 40.23 -> 40.51 | No clear effect |
| add Spatial Geometric Aligner (ablation) | InstaDrive · nuScenes val (T+I)2V | NDS: 37.31 -> 40.51 | Improved |
| add Spatial Geometric Aligner (ablation) | InstaDrive · nuScenes val (T+I)2V | IDS: 723 -> 532 | Improved |
| add Spatial Geometric Aligner (ablation) | InstaDrive · nuScenes val (T+I)2V | FVD: 41.21 -> 38.06 | Improved |
| use box projection instead of box coordinates as control (ablation) | InstaDrive · nuScenes val (T+I)2V | NDS: 38.22 -> 40.51 | Improved |
| add depth-order cues (ablation) | InstaDrive · nuScenes val (T+I)2V | NDS: 39.13 -> 40.51 | Improved |
| add generated data to real data when training StreamPETR | StreamPETR · nuScenes detection | NDS: 48.3 -> 51.9 | Improved |
| add generated data to real data when training StreamPETR | StreamPETR · nuScenes detection | mAP: 38.3 -> 42.0 | Improved |
| add InstaDrive generated data to real data for tracking | StreamPETR · nuScenes MOT | IDS: 687 -> 532 | Improved |
Later papers that compared with this method
1 later accepted paper reports InstaDrive as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- ConsisDrive: Identity-Preserving Driving World Models for Video Generation by Instance Mask (ICLR 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 5 of 5 on most shared results.
- DriveDreamer2 — DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation (AAAI 2025) · better on 4 of 4 shared results
- MagicDrive-V2 — MagicDrive-V2: High-Resolution Long Video Generation for Autonomous Driving with Adaptive Control (ICCV 2025) · better on 3 of 3 shared results
- Panacea — Panacea: Panoramic and Controllable Video Generation for Autonomous Driving (CVPR 2024) · better on 3 of 3 shared results
- DriveScape — DriveScape: High-Resolution Driving Video Generation by Multi-View Feature Fusion (CVPR 2025) · better on 2 of 2 shared results
- DrivingDiffusion — DrivingDiffusion: Layout-Guided Multi-View Driving Scenarios Video Generation with Latent Diffusion Model (ECCV 2024) · 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 0.89)
- SceneCrafter: Controllable Multi-View Driving Scene Editing (CVPR 2025 · similarity 0.88)
- Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (CVPR 2024 · similarity 0.87)
- DrivingSphere: Building a High-fidelity 4D World for Closed-loop Simulation (CVPR 2025 · similarity 0.87)
- X-Scene: Large-Scale Driving Scene Generation with High Fidelity and Flexible Controllability (NeurIPS 2025 · similarity 0.87)
- DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation (AAAI 2025 · similarity 0.87)
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.
