Breaking the Vicious Cycle: Coherent 3D Gaussian Splatting from Sparse and Motion-Blurred Views

TL;DR

This paper introduces CoherentGS, a novel framework that leverages dual priors from pre-trained generative models to achieve high-fidelity 3D reconstruction from sparse, motion-blurred images, overcoming limitations of existing methods.

Contribution

The paper proposes a dual-prior strategy combining deblurring and geometric priors, along with innovative techniques like camera exploration and depth regularization, to improve 3D reconstruction from degraded images.

Findings

Outperforms existing methods on synthetic and real scenes

Achieves high-quality 3D reconstructions from as few as 3 input views

Sets new state-of-the-art results in the field

Abstract

3D Gaussian Splatting (3DGS) has emerged as a state-of-the-art method for novel view synthesis. However, its performance heavily relies on dense, high-quality input imagery, an assumption that is often violated in real-world applications, where data is typically sparse and motion-blurred. These two issues create a vicious cycle: sparse views ignore the multi-view constraints necessary to resolve motion blur, while motion blur erases high-frequency details crucial for aligning the limited views. Thus, reconstruction often fails catastrophically, with fragmented views and a low-frequency bias. To break this cycle, we introduce CoherentGS, a novel framework for high-fidelity 3D reconstruction from sparse and blurry images. Our key insight is to address these compound degradations using a dual-prior strategy. Specifically, we combine two pre-trained generative models: a specialized deblurring network for restoring sharp details and providing photometric guidance, and a diffusion model that offers geometric priors to fill in unobserved regions of the scene. This dual-prior strategy is supported by several key techniques, including a consistency-guided camera exploration module that adaptively guides the generative process, and a depth regularization loss that ensures geometric plausibility. We evaluate CoherentGS through both quantitative and qualitative experiments on synthetic and real-world scenes, using as few as 3, 6, and 9 input views. Our results demonstrate that CoherentGS significantly outperforms existing methods, setting a new state-of-the-art for this challenging task. The code and video demos are available at https://potatobigroom.github.io/CoherentGS/.