Research map / 3D vision / Gaussian splatting
Catalyst4D: High-Fidelity 3D-to-4D Scene Editing via Dynamic Propagation
Shifeng Chen, Yihui Li, Jun Liao, Hongyu Yang, Di Huang
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 Anchor-based Motion Guidance (AMG) (ablation) | Catalyst4D · DyNeRF/MeetRoom | CLIP similarity: 0.245 -> 0.252 | Improved |
| add Anchor-based Motion Guidance (AMG) (ablation) | Catalyst4D · DyNeRF/MeetRoom | VBench consistency: 0.966 -> 0.971 | Improved |
| add CUAR appearance refinement (ablation) | Catalyst4D · DyNeRF/MeetRoom | CLIP similarity: 0.248 -> 0.252 | Improved |
| add CUAR appearance refinement (ablation) | Catalyst4D · DyNeRF/MeetRoom | VBench consistency: 0.969 -> 0.971 | Improved |
| add AMG and CUAR to deformation-network baseline (ablation) | Catalyst4D · 16 prompts | EditScore: 2.713 -> 7.375 | Improved |
| add AMG and CUAR to deformation-network baseline (ablation) | Catalyst4D · 16 prompts | VE-Bench: 0.059 -> 1.080 | Improved |
| add CUAR on top of AMG (ablation) | Catalyst4D · 16 prompts | EditScore: 5.112 -> 7.375 | Improved |
| add AMG on top of CUAR (ablation) | Catalyst4D · 16 prompts | VE-Bench: 0.580 -> 1.080 | Improved |
| use Catalyst4D instead of CTRL-D | Catalyst4D · 16 prompts | EditScore: 4.326 -> 7.375 | Improved |
| use Catalyst4D instead of Instruct-4DGS | Catalyst4D · Sear-steak | CLIP similarity: 0.220 -> 0.252 | Improved |
| use Catalyst4D instead of CTRL-D | Catalyst4D · Sear-steak | VBench consistency: 0.985 -> 0.983 | No clear effect |
| use Catalyst4D instead of Instruct 4D-to-4D for training time | Catalyst4D · Sear-steak | training time: 2 h -> 50 min | Improved |
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.
- CTRL-D — CTRL-D: Controllable Dynamic 3D Scene Editing with Personalized 2D Diffusion (CVPR 2025) · better on 7 of 8 shared results
- Instruct 4D-to-4D — Instruct 4D-to-4D: Editing 4D Scenes as Pseudo-3D Scenes Using 2D Diffusion (CVPR 2024) · better on 4 of 4 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
- Instruct 4D-to-4D: Editing 4D Scenes as Pseudo-3D Scenes Using 2D Diffusion (CVPR 2024 · similarity 0.92)
- Efficient Dynamic Scene Editing via 4D Gaussian-based Static-Dynamic Separation (CVPR 2025 · similarity 0.91)
- Control4D: Efficient 4D Portrait Editing with Text (CVPR 2024 · similarity 0.89)
- CTRL-D: Controllable Dynamic 3D Scene Editing with Personalized 2D Diffusion (CVPR 2025 · similarity 0.89)
- Motion4D: Learning 3D-Consistent Motion and Semantics for 4D Scene Understanding (NeurIPS 2025 · similarity 0.89)
- EditSplat: Multi-View Fusion and Attention-Guided Optimization for View-Consistent 3D Scene Editing with 3D Gaussian Splatting (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.
