ArtifactWorld: Scaling 3D Gaussian Splatting Artifact Restoration via Video Generation Models

TL;DR

ArtifactWorld introduces a comprehensive framework combining data expansion and a dual-model paradigm to effectively restore 3D Gaussian Splatting artifacts, significantly improving sparse-view rendering quality.

Contribution

The paper presents a novel artifact restoration method using a large-scale dataset, a unified video diffusion model, and an artifact-aware fusion mechanism for 3DGS.

Findings

Achieves state-of-the-art results in sparse novel view synthesis

Demonstrates robust 3D reconstruction under artifact conditions

Constructed a large dataset of 107.5K paired video clips for training

Abstract

3D Gaussian Splatting (3DGS) delivers high-fidelity real-time rendering but suffers from geometric and photometric degradations under sparse-view constraints. Current generative restoration approaches are often limited by insufficient temporal coherence, a lack of explicit spatial constraints, and a lack of large-scale training data, resulting in multi-view inconsistencies, erroneous geometric hallucinations, and limited generalization to diverse real-world artifact distributions. In this paper, we present ArtifactWorld, a framework that resolves 3DGS artifact repair through systematic data expansion and a homogeneous dual-model paradigm. To address the data bottleneck, we establish a fine-grained phenomenological taxonomy of 3DGS artifacts and construct a comprehensive training set of 107.5K diverse paired video clips to enhance model robustness. Architecturally, we unify the restoration process within a video diffusion backbone, utilizing an isomorphic predictor to localize structural defects via an artifact heatmap. This heatmap then guides the restoration through an Artifact-Aware Triplet Fusion mechanism, enabling precise, intensity-guided spatio-temporal repair within native self-attention. Extensive experiments demonstrate that ArtifactWorld achieves state-of-the-art performance in sparse novel view synthesis and robust 3D reconstruction. Code and dataset will be made public.