Catalyst4D: High-Fidelity 3D-to-4D Scene Editing via Dynamic Propagation

2026-03-13

AI Summary

This work addresses key challenges in dynamic scene editing—such as motion artifacts, temporal flickering, and inconsistent style propagation—by proposing a spatiotemporally coherent, high-fidelity editing framework that transfers high-quality 3D edits to dynamic 4D Gaussian scenes. The approach introduces two core innovations: an anchor-based motion guidance mechanism that effectively mitigates cross-region interference and motion drift, and a color-uncertainty-guided appearance optimization strategy that enhances visual consistency across time. By integrating 3D Gaussian representations, optimal transport, anchor correspondence construction, and selective appearance refinement, the method achieves significant improvements over existing techniques in both visual fidelity and motion coherence, enabling stable and high-quality editing of dynamic scenes.

Technology Category

• Computer Vision: Motion & Tracking
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Data transparency and provenance
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Recent advances in 3D scene editing using NeRF and 3DGS enable high-quality static scene editing. In contrast, dynamic scene editing remains challenging, as methods that directly extend 2D diffusion models to 4D often produce motion artifacts, temporal flickering, and inconsistent style propagation. We introduce Catalyst4D, a framework that transfers high-quality 3D edits to dynamic 4D Gaussian scenes while maintaining spatial and temporal coherence. At its core, Anchor-based Motion Guidance (AMG) builds a set of structurally stable and spatially representative anchors from both original and edited Gaussians. These anchors serve as robust region-level references, and their correspondences are established via optimal transport to enable consistent deformation propagation without cross-region interference or motion drift. Complementarily, Color Uncertainty-guided Appearance Refinement (CUAR) preserves temporal appearance consistency by estimating per-Gaussian color uncertainty and selectively refining regions prone to occlusion-induced artifacts. Extensive experiments demonstrate that Catalyst4D achieves temporally stable, high-fidelity dynamic scene editing and outperforms existing methods in both visual quality and motion coherence.

Problem

• 4D scene editing
• temporal coherence
• motion artifacts
• dynamic scene
• spatial consistency

Innovation

• 4D scene editing
• Anchor-based Motion Guidance
• Color Uncertainty-guided Appearance Refinement
• dynamic Gaussian splatting
• temporal coherence