2Xplat: Two Experts Are Better Than One Generalist

TL;DR

2Xplat introduces a modular two-expert framework for 3D Gaussian Splatting that separates geometry estimation from appearance synthesis, leading to faster training and superior performance compared to monolithic models.

Contribution

The paper proposes a novel two-expert architecture for pose-free 3D Gaussian Splatting, explicitly decoupling geometry and appearance modeling for improved results.

Findings

Outperforms prior pose-free methods in fewer than 5K iterations

Achieves performance comparable to state-of-the-art pose-based methods

Challenges the all-in-one paradigm with a modular approach

Abstract

Pose-free feed-forward 3D Gaussian Splatting (3DGS) has opened a new frontier for rapid 3D modeling, enabling high-quality Gaussian representations to be generated from uncalibrated multi-view images in a single forward pass. The dominant approach in this space adopts unified monolithic architectures, often built on geometry-centric 3D foundation models, to jointly estimate camera poses and synthesize 3DGS representations within a single network. While architecturally streamlined, such "all-in-one" designs may be suboptimal for high-fidelity 3DGS generation, as they entangle geometric reasoning and appearance modeling within a shared representation. In this work, we introduce 2Xplat, a pose-free feed-forward 3DGS framework based on a two-expert design that explicitly separates geometry estimation from Gaussian generation. A dedicated geometry expert first predicts camera poses, which are then explicitly passed to a powerful appearance expert that synthesizes 3D Gaussians. Despite its conceptual simplicity, being largely underexplored in prior works, the proposed approach proves highly effective. In fewer than 5K training iterations, the proposed two-experts pipeline substantially outperforms prior pose-free feed-forward 3DGS approaches and achieves performance on par with state-of-the-art posed methods. These results challenge the prevailing unified paradigm and suggest the potential advantages of modular design principles for complex 3D geometric estimation and appearance synthesis tasks.