AI Paper Atlas

Research map / Video / Video generation

3D Scene Prompting for Scene-Consistent Camera-Controllable Video Generation

JoungBin Lee, Jaewoo Jung, Jisang Han, Takuya Narihira, Kazumi Fukuda, Junyoung Seo, Sunghwan Hong, Yuki Mitsufuji, Seungryong Kim

ICLR 2026 · Video › Video generation · approach: Trajectory-guided motion control

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 dynamic masking before building the static point cloud (ablation)3DScenePrompt · DynPose-100KPSNR: 12.2304 -> 13.0468Improved
add spatial prompts (n=7 projected views) to temporal-only conditioning (ablation)3DScenePrompt · DynPose-100KPSNR: 11.9555 -> 13.0468Improved
add spatial prompts (n=7) to temporal-only conditioning (ablation)3DScenePrompt · DynPose-100K camera controlmRotErr: 3.4142 -> 2.3772Improved
increase spatial frames n beyond 7 (to all frames L) (ablation)3DScenePrompt · DynPose-100KPSNR / MEt3R: PSNR 13.0468 -> 13.0206; MEt3R 0.124189 -> 0.123507No clear effect
increase temporal window from w=1 to w=9 (ablation)3DScenePrompt · DynPose-100K camera controlmCamMC: 8.9785 -> 8.6352No clear effect
increase temporal window from w=1 to w=9 (ablation)3DScenePrompt · DynPose-100KMotion Smooth: 0.9253 -> 0.9335Improved
add SAM2 propagation to L1 flow mask (ablation)dynamic masking pipeline · DynPose-100K subsetPSNR: 12.1083 -> 13.5957Improved
add backward point tracking before SAM2 (ablation)dynamic masking pipeline · DynPose-100K subsetMEt3R: 0.157610 -> 0.122859Improved
add dynamic mask pipeline (ablation)3DScenePrompt · DynPose-100K subsetMEt3R: 0.104002 -> 0.122859Mixed
replace MegaSAM with DepthAnything v3 for 3D memory (ablation)3DScenePrompt · DynPose-100KPSNR: 13.0468 -> 13.4534Improved
use spatial 3D memory instead of DFoT history guidancescene-consistent generation · RealEstate10KMEt3R: 0.181164 -> 0.040843Improved

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