MoRE: 3D Visual Geometry Reconstruction Meets Mixture-of-Experts

TL;DR

MoRE is a scalable 3D visual foundation model using a Mixture-of-Experts architecture that improves geometric reconstruction robustness and accuracy across diverse tasks and real-world conditions.

Contribution

The paper introduces MoRE, a novel dense 3D model leveraging MoE architecture, confidence-based refinement, and semantic integration for enhanced scalability and task adaptability.

Findings

Achieves state-of-the-art results on multiple benchmarks.

Supports diverse 3D tasks without additional computation.

Improves robustness and accuracy in real-world scenarios.

Abstract

Recent advances in language and vision have demonstrated that scaling up model capacity consistently improves performance across diverse tasks. In 3D visual geometry reconstruction, large-scale training has likewise proven effective for learning versatile representations. However, further scaling of 3D models is challenging due to the complexity of geometric supervision and the diversity of 3D data. To overcome these limitations, we propose MoRE, a dense 3D visual foundation model based on a Mixture-of-Experts (MoE) architecture that dynamically routes features to task-specific experts, allowing them to specialize in complementary data aspects and enhance both scalability and adaptability. Aiming to improve robustness under real-world conditions, MoRE incorporates a confidence-based depth refinement module that stabilizes and refines geometric estimation. In addition, it integrates dense semantic features with globally aligned 3D backbone representations for high-fidelity surface normal prediction. MoRE is further optimized with tailored loss functions to ensure robust learning across diverse inputs and multiple geometric tasks. Extensive experiments demonstrate that MoRE achieves state-of-the-art performance across multiple benchmarks and supports effective downstream applications without extra computation.