Research map / 3D vision / 3D generation
HorizonForge: Driving Scene Editing with Any Trajectories and Any Vehicles
Yifan Wang, Francesco Pittaluga, Zaid Tasneem, Chenyu You, Manmohan Chandraker, Ziyu Jiang
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 |
|---|---|---|---|
| use Gaussian-mesh conditioning instead of colored point clouds (ablation) | HorizonForge · HorizonSuite ego speed change | FID: 62.16 -> 39.92 | Improved |
| use Gaussian-mesh conditioning instead of bounding boxes (ablation) | HorizonForge · HorizonSuite ego speed change | FID: 108.54 -> 39.92 | Improved |
| use video diffusion instead of single-frame Image DM (ablation) | HorizonForge · HorizonSuite ego speed change | FID: 82.19 -> 39.92 | Improved |
| use video diffusion instead of Image DM (ablation) | HorizonForge · HorizonSuite overall | FID: 75.83 -> 33.19 | Improved |
| use Gaussian-mesh conditioning instead of point clouds (ablation) | HorizonForge · HorizonSuite overall | FID: 54.14 -> 33.19 | Improved |
| use HorizonForge instead of OmniRe on ego direction change | OmniRe · HorizonSuite ego direction change | FID: 80.60 -> 144.83 | Hurt |
| use HorizonForge instead of OmniRe on insertion | OmniRe · HorizonSuite vehicle insertion | FID: 182.29 -> 117.46 | Improved |
| use HorizonForge instead of StreetCrafter overall | StreetCrafter · HorizonSuite overall | FID: 91.16 -> 33.19 | Improved |
| add edited scenes with inserted vehicles to detector training | 3D detector · Downstream detection IoU 0.5 | AP: 0.0533 -> 0.0635 | Improved |
| add edited scenes with removed vehicles to detector training | 3D detector · Downstream detection IoU 0.7 | AP: 0.0103 -> 0.0134 | Improved |
| use HorizonForge instead of ReconDreamer for 6m lane shift | ReconDreamer · Lane shift 6m | FID: 149.19 -> 53.61 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- OmniRe — OmniRe: Omni Urban Scene Reconstruction (ICLR 2025) · better on 6 of 7 shared results
- ReconDreamer — ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration (CVPR 2025) · better on 1 of 1 shared result
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
- SIMSplat: Language-Aligned 4D Gaussian Splatting for Driving Scenario Generation (ECCV 2026 · similarity 0.92)
- SceneCrafter: Controllable Multi-View Driving Scene Editing (CVPR 2025 · similarity 0.90)
- X-Scene: Large-Scale Driving Scene Generation with High Fidelity and Flexible Controllability (NeurIPS 2025 · similarity 0.89)
- DriveEditor: A Unified 3D Information-Guided Framework for Controllable Object Editing in Driving Scenes (AAAI 2025 · similarity 0.89)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (CVPR 2025 · similarity 0.87)
- 3DitScene: Editing Any Scene via Language-guided Disentangled Gaussian Splatting (ICLR 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.
