Research map / 3D vision / 3D generation
Now You Shake Me: Towards Automatic 4D Cinema
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| add attention pooling (ablation) | trimmed effect recognition model · Movie4D test set | E: 41.5 -> 43.7 | Helped |
| add attention pooling (ablation) | trimmed effect recognition model · Movie4D test set | DA-PR: 5.3 -> 4.6 | Hurt |
| remove visual features (ablation) | trimmed effect recognition model · Movie4D test set | E: 41.5 -> 26.9 | Hurt |
| remove optical-flow features (ablation) | trimmed effect recognition model · Movie4D test set | E: 41.5 -> 32.1 | Hurt |
| remove audio features (ablation) | trimmed effect recognition model · Movie4D test set | E: 41.5 -> 40.7 | Hurt |
| remove object-detection features (ablation) | trimmed effect recognition model · Movie4D test set | E: 41.5 -> 36.7 | Hurt |
| add all CRF pairwise potentials (ablation) | unary model for untrimmed effects, camera POV · Movie4D test set | Exist AP: 55.4 -> 57.3 | Helped |
| add all CRF pairwise potentials (ablation) | unary model for untrimmed effects, camera POV · Movie4D test set | Effect ACC: 43.6 -> 52.4 | Helped |
| add the CRF to group predictions into time intervals (ablation) | unary model for untrimmed effects, camera POV · Movie4D test set | F1: 21.1 -> 29.5 | Helped |
| add the CRF to group predictions into time intervals (ablation) | unary model for untrimmed effects, camera and tracks · Movie4D test set | F1: 20.5 -> 21.1 | Helped |
Closest papers by meaning
- Motion4D: Learning 3D-Consistent Motion and Semantics for 4D Scene Understanding (NeurIPS 2025 · similarity 0.84)
- CineTechBench: A Benchmark for Cinematographic Technique Understanding and Generation (NeurIPS 2025 · similarity 0.83)
- Feature4X: Bridging Any Monocular Video to 4D Agentic AI with Versatile Gaussian Feature Fields (CVPR 2025 · similarity 0.83)
- Visual Objectification in Films: Towards a New AI Task for Video Interpretation (CVPR 2024 · similarity 0.83)
- ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models (NeurIPS 2025 · similarity 0.82)
- MovieBench: A Hierarchical Movie Level Dataset for Long Video Generation (CVPR 2025 · similarity 0.82)
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.