LU-NeRF: Scene and Pose Estimation by Synchronizing Local Unposed NeRFs

TL;DR

LU-NeRF introduces a novel method for jointly estimating camera poses and scene representations in NeRF models, operating under relaxed assumptions and outperforming prior methods in unposed scenarios.

Contribution

The paper presents LU-NeRF, a local-to-global approach that estimates poses and scenes without restrictive priors, enabling broader application of NeRF in unstructured environments.

Findings

Outperforms prior unposed NeRF methods in pose and scene estimation.

Operates effectively in general SE(3) pose settings.

Complements feature-based SfM pipelines, especially on low-texture images.

Abstract

A critical obstacle preventing NeRF models from being deployed broadly in the wild is their reliance on accurate camera poses. Consequently, there is growing interest in extending NeRF models to jointly optimize camera poses and scene representation, which offers an alternative to off-the-shelf SfM pipelines which have well-understood failure modes. Existing approaches for unposed NeRF operate under limited assumptions, such as a prior pose distribution or coarse pose initialization, making them less effective in a general setting. In this work, we propose a novel approach, LU-NeRF, that jointly estimates camera poses and neural radiance fields with relaxed assumptions on pose configuration. Our approach operates in a local-to-global manner, where we first optimize over local subsets of the data, dubbed mini-scenes. LU-NeRF estimates local pose and geometry for this challenging few-shot task. The mini-scene poses are brought into a global reference frame through a robust pose synchronization step, where a final global optimization of pose and scene can be performed. We show our LU-NeRF pipeline outperforms prior attempts at unposed NeRF without making restrictive assumptions on the pose prior. This allows us to operate in the general SE(3) pose setting, unlike the baselines. Our results also indicate our model can be complementary to feature-based SfM pipelines as it compares favorably to COLMAP on low-texture and low-resolution images.