AI Paper Atlas

Research map / Video / Video generation

GeoFlow: Efficient Driving Video Generation via Geometry-Aligned Priors

Jiazheng Liu, Hang Li, jiawei zhang, Jiahe Li, Xiaohan Yu, Shengyin Fan, Jin Zheng, Xiao Bai

ECCV 2026 · Video › Video generation · approach: Autoregressive long video generation · 0 citations (OpenAlex)

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
replace Gaussian source with geometry-aligned prior, 5 stepsOpenDWM · NuScenes val, 5 stepsFVD: 123.5 -> 49.2Improved
replace Gaussian source with geometry-aligned prior, 8 steps vs baseline 40 stepsOpenDWM · NuScenes valFVD: 38.8 -> 38.6Improved
add GeoFlow prior to temporal-VAE baseOpenDWM-tvae · NuScenes val, 5 stepsFVD: 204.8 -> 76.7Improved
add GeoFlow prior to spatial-VAE baseOpenDWM-vae · NuScenes val, 5 stepsFVD: 121.7 -> 49.2Improved
add GeoFlow prior to stronger base modelUniMLVG · NuScenes val, 5 stepsFVD: 72.6 -> 44.6Improved
GeoFlow at full 40 steps vs baselineOpenDWM · NuScenes val, 40 stepsFID/FVD: FID 6.8 -> 6.9, FVD 38.8 -> 34.0Mixed
add occlusion mask M_occ (ablation)GeoFlow noise mask · NuScenes val, 10 stepsFVD: 43.6 -> 40.0Improved
add depth uncertainty mask M_unc on top of M_occ (ablation)GeoFlow noise mask · NuScenes val, 10 stepsFVD: 40.0 -> 40.2No clear effect
add dynamic-object mask M_dyn (ablation)GeoFlow noise mask · NuScenes val, 10 stepsFVD: 40.2 -> 35.0Improved
use full spatially-adaptive noise mask instead of global uniform noise (ablation)GeoFlow · NuScenes valFVDImproved
inject any noise instead of naive warping without noise (ablation)GeoFlow · NuScenes valFVDImproved
swap depth model to MapAnything-v1.1 zero-shot (ablation)GeoFlow · NuScenes val, 10 stepsFVD: 35.0 -> 33.1Improved

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.

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