Alias-free 4D Gaussian Splatting

2025-11-23

AI Summary

Existing 4D Gaussian splatting-based methods for dynamic scene reconstruction suffer from severe aliasing artifacts during zooming or multi-resolution rendering, due to mismatch between Gaussian scale and camera sampling frequency. This work proposes a scale-adaptive 4D Gaussian representation: first, leveraging anti-aliasing theory, we derive a closed-form expression for the maximum safe sampling frequency, explicitly characterizing the scale–frequency constraint; second, we design a differentiable 4D scale-adaptive filter and introduce a scale-aware loss function to jointly optimize Gaussian anisotropic scales and density distributions. The method effectively suppresses high-frequency aliasing, significantly reduces redundant Gaussian primitives, and enables aliasing-free, high-fidelity, multi-resolution-compatible rendering for both monocular and multi-view video reconstruction. Quantitatively, it improves PSNR and SSIM while reducing memory overhead.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Existing dynamic scene reconstruction methods based on Gaussian Splatting enable real-time rendering and generate realistic images. However, adjusting the camera's focal length or the distance between Gaussian primitives and the camera to modify rendering resolution often introduces strong artifacts, stemming from the frequency constraints of 4D Gaussians and Gaussian scale mismatch induced by the 2D dilated filter. To address this, we derive a maximum sampling frequency formulation for 4D Gaussian Splatting and introduce a 4D scale-adaptive filter and scale loss, which flexibly regulates the sampling frequency of 4D Gaussian Splatting. Our approach eliminates high-frequency artifacts under increased rendering frequencies while effectively reducing redundant Gaussians in multi-view video reconstruction. We validate the proposed method through monocular and multi-view video reconstruction experiments.Ours project page: https://4d-alias-free.github.io/4D-Alias-free/

Problem

• Eliminates high-frequency artifacts in dynamic scene reconstruction
• Addresses Gaussian scale mismatch from 2D dilated filters
• Reduces redundant Gaussians in multi-view video reconstruction

Innovation

• Derived maximum sampling frequency for 4D Gaussian Splatting
• Introduced 4D scale-adaptive filter for flexible sampling
• Proposed scale loss to reduce redundant Gaussians effectively