Research map / Video / Video generation
PhysChoreo: Physics-Controllable Video Generation with Part-Aware Semantic Grounding
Haoze Zhang, Tianyu Huang, Zichen Wan, Xiaowei Jin, Hongzhi Zhang, Hui Li, Wangmeng Zuo
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 |
|---|---|---|---|
| add cross-attention text conditioning to plain transformer (ablation) | PhysChoreo predictor · Ablation: model design | Material accuracy / total error: 0.5507 -> 0.7393; 3.4230 -> 0.9789 | Improved |
| add dual-stage (global then part) cross-attention (ablation) | PhysChoreo predictor · Ablation: model design | Material accuracy / total error: 0.7393 -> 0.8077; 0.9789 -> 0.5801 | Improved |
| add soft assignment (ablation) | PhysChoreo predictor · Ablation: model design | Material accuracy / total error: 0.8077 -> 0.8372; 0.5801 -> 0.5046 | Improved |
| add segmentation prior (ablation) | PhysChoreo predictor · Ablation: model design | Material accuracy / total error: 0.8372 -> 0.8605; 0.5046 -> 0.3318 | Improved |
| add soft assignment (ablation) | PhysChoreo predictor · Ablation: model design | Early-stopping iterations: 36 -> 26 | Improved |
| add segmentation prior (ablation) | PhysChoreo predictor · Ablation: model design | Early-stopping iterations: 26 -> 41 | Hurt |
| add L_assign loss (ablation) | PhysChoreo predictor · Ablation: loss components | Material accuracy / total error: 0.8534 -> 0.8605; 0.3753 -> 0.3318 | Improved |
| add L_assign loss (ablation) | PhysChoreo predictor · Ablation: loss components | Early-stopping iterations: 50 -> 41 | Improved |
| add L_smooth loss (ablation) | PhysChoreo predictor · Ablation: loss components | Material accuracy / total error: 0.8578 -> 0.8605; 0.3579 -> 0.3318 | Improved |
| add L_con loss (ablation) | PhysChoreo predictor · Ablation: loss components | Material accuracy: 0.8310 -> 0.8605 | Improved |
| add L_con loss (ablation) | PhysChoreo predictor · Ablation: loss components | Total error: 0.3451 -> 0.3318 | No clear effect |
| use part-level predicted physics instead of global properties (ablation) | PhysChoreo pipeline · Video generation | Part-level dynamics | Improved |
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.
- PhysCtrl — PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation (NeurIPS 2025) · better on 8 of 8 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
- PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation (NeurIPS 2025 · similarity 0.92)
- VideoPhy: Evaluating Physical Commonsense for Video Generation (ICLR 2025 · similarity 0.89)
- Hierarchical Fine-grained Preference Optimization for Physically Plausible Video Generation (NeurIPS 2025 · similarity 0.88)
- DreamPhysics: Learning Physics-Based 3D Dynamics with Video Diffusion Priors (AAAI 2025 · similarity 0.88)
- PhysGen3D: Crafting a Miniature Interactive World from a Single Image (CVPR 2025 · similarity 0.88)
- PhysAnimator: Physics-Guided Generative Cartoon Animation (CVPR 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.
