ProSplat: Improved Feed-Forward 3D Gaussian Splatting for Wide-Baseline Sparse Views

TL;DR

ProSplat is a novel two-stage framework that significantly improves wide-baseline 3D view synthesis by enhancing Gaussian primitives with a diffusion-based model and geometric consistency techniques.

Contribution

It introduces a two-stage feed-forward approach with a diffusion-based enhancement and novel geometric consistency methods for wide-baseline view synthesis.

Findings

Achieves 1 dB higher PSNR than state-of-the-art methods.

Effectively handles wide-baseline scenarios with improved texture and geometric consistency.

Demonstrates superior performance on RealEstate10K and DL3DV-10K datasets.

Abstract

Feed-forward 3D Gaussian Splatting (3DGS) has recently demonstrated promising results for novel view synthesis (NVS) from sparse input views, particularly under narrow-baseline conditions. However, its performance significantly degrades in wide-baseline scenarios due to limited texture details and geometric inconsistencies across views. To address these challenges, in this paper, we propose ProSplat, a two-stage feed-forward framework designed for high-fidelity rendering under wide-baseline conditions. The first stage involves generating 3D Gaussian primitives via a 3DGS generator. In the second stage, rendered views from these primitives are enhanced through an improvement model. Specifically, this improvement model is based on a one-step diffusion model, further optimized by our proposed Maximum Overlap Reference view Injection (MORI) and Distance-Weighted Epipolar Attention (DWEA). MORI supplements missing texture and color by strategically selecting a reference view with maximum viewpoint overlap, while DWEA enforces geometric consistency using epipolar constraints. Additionally, we introduce a divide-and-conquer training strategy that aligns data distributions between the two stages through joint optimization. We evaluate ProSplat on the RealEstate10K and DL3DV-10K datasets under wide-baseline settings. Experimental results demonstrate that ProSplat achieves an average improvement of 1 dB in PSNR compared to recent SOTA methods.