AI Paper Atlas

Research map / Video / Video generation

RecEdit-Drive: 3D Reconstruction-Guided Spatiotemporal Video Editing for Autonomous Driving Scenes

Yipeng Wu, Xin Wang, Chenghan Yang, Chong Wang, Dongdong Wu, Wanchao Su, Hengshuang Zhao, Wei Feng, Kairui Yang, Di Lin

CVPR 2026 · Video › Video generation · approach: World-model video generation

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 changedWhereMeasuredEffect
add Spatiotemporal Collaborative Modeling (SCM) to SFW (ablation)RecEdit-Drive · nuScenes editing ablationFVD: 18.62 -> 15.02Improved
add Noise Replace to SFW + SCM (ablation)RecEdit-Drive · nuScenes editing ablationFVD: 15.02 -> 14.38Improved
add Noise Replace to SFW only (ablation)RecEdit-Drive · nuScenes editing ablationFVD: 18.62 -> 18.44No clear effect
add SCM and Noise Replace on top of SFW (ablation)RecEdit-Drive · nuScenes editing ablationPSNR: 30.15 -> 31.46Improved
add SCM and Noise Replace on top of SFW (ablation)RecEdit-Drive · nuScenes editing ablationmRecall: 0.960 -> 0.964No clear effect
use SFW instead of VGGT-rendered 3D prior (ablation)RecEdit-Drive · nuScenes 3D prior ablationFVD: 18.53 -> 14.38Improved
use SFW instead of single-view SV3D features (ablation)RecEdit-Drive · nuScenes 3D prior ablationFVD: 18.21 -> 14.38Improved
use SFW instead of SV3D prior for spatial control (ablation)RecEdit-Drive · nuScenes 3D prior ablationmATE: 0.5891 -> 0.5757Improved
augment StreamPETR training with generated repositioning and replacement dataStreamPETR · nuScenes detection, 50% training subsetmAP: 0.4796 -> 0.4888Improved
augment StreamPETR training with generated repositioning and replacement dataStreamPETR · nuScenes detection, 50% training subsetNDS: 0.5617 -> 0.5905Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy