Research map / Video / Video generation
CineScene: Implicit 3D as Effective Scene Representation for Cinematic Video Generation
Kaiyi Huang, Yukun Huang, Yu Li, Jianhong Bai, Xintao Wang, Zinan Lin, Xuefei Ning, Jiwen Yu, Yu Wang, Xihui Liu
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 |
|---|---|---|---|
| inject implicit 3D via context conditioning instead of VGGT loss guidance (ablation) | CineScene · Scene-Decoupled test | PSNR: 13.9996 -> 14.5094 | Improved |
| add implicit 3D (VGGT image+camera features) vs none (ablation) | CineScene · Scene-Decoupled test | PSNR: 13.7582 -> 14.5094 | Improved |
| use only image feature instead of fused image+camera (ablation) | CineScene · Scene-Decoupled test | PSNR: 14.5094 -> 13.9623 | Hurt |
| use only camera feature instead of fused (ablation) | CineScene · Scene-Decoupled test | Mat. Pix.: 4617.51 -> 4498.83 | Hurt |
| shuffle scene context images during training vs ordered (ablation) | CineScene · Scene-Decoupled test | PSNR: 14.0174 -> 14.5094 | Improved |
| shuffle vs ordered on Mat. Pix. (ablation) | CineScene · Scene-Decoupled test | Mat. Pix.: 4673.67 -> 4617.51 | Hurt |
| shuffle vs progressive training (ablation) | CineScene · Scene-Decoupled test | RotErr: 2.5757 -> 2.6825 | Mixed |
| use CineScene implicit 3D context instead of explicit 3D guidance (Gen3C) | CineScene · Scene-Decoupled test | PSNR: 11.6255 -> 14.5094 | Improved |
| use CineScene instead of FramePack on text alignment | CineScene · Scene-Decoupled test | CLIP-T: 0.3269 -> 0.3212 | Hurt |
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.
- RecamMaster — ReCamMaster: Camera-Controlled Generative Rendering from A Single Video (ICCV 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
- CineTechBench: A Benchmark for Cinematographic Technique Understanding and Generation (NeurIPS 2025 · similarity 0.89)
- Unveiling the Value of Motion for Cinematic Camera Trajectories (NeurIPS 2026 · similarity 0.88)
- CineOrchestra: Unified Entity-Centric Conditioning for Cinematic Video Generation (NeurIPS 2026 · similarity 0.88)
- Auteur: Language-Driven Cinematographic Framing for Human-Centric Video Generation (NeurIPS 2026 · similarity 0.88)
- A 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale (NeurIPS 2026 · similarity 0.88)
- Video Perception Models for 3D Scene Synthesis (NeurIPS 2025 · similarity 0.88)
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.
