PCM-NeRF: Probabilistic Camera Modeling for Neural Radiance Fields under Pose Uncertainty

TL;DR

PCM-NeRF introduces a probabilistic approach to neural radiance fields that models camera pose uncertainty, improving reconstruction quality in scenes with pose errors by adaptively damping updates based on view confidence.

Contribution

It presents a novel probabilistic framework that incorporates learnable pose uncertainty into neural surface reconstruction, enhancing robustness against pose inaccuracies.

Findings

Outperforms state-of-the-art methods in Chamfer Distance and F-Score.

Effectively handles scenes with severe pose outliers.

Improves reconstruction of geometrically complex structures.

Abstract

Neural surface reconstruction methods typically treat camera poses as fixed values, assuming perfect accuracy from Structure-from-Motion (SfM) systems. This assumption breaks down with imperfect pose estimates, leading to distorted or incomplete reconstructions. We present PCM-NeRF, a probabilistic framework that augments neural surface reconstruction with per-camera learnable uncertainty, built on top of SG-NeRF. Rather than treating all cameras equally throughout optimization, we represent each pose as a distribution with a learnable mean and variance, initialized from SfM correspondence quality. An uncertainty regularization loss couples the learned variance to view confidence, and the resulting uncertainty directly modulates the effective pose learning rate: uncertain cameras receive damped gradient updates, preventing poorly initialized views from corrupting the reconstruction. This lightweight mechanism requires no changes to the rendering pipeline and adds negligible overhead. Experiments on challenging scenes with severe pose outliers demonstrate that PCM-NeRF consistently outperforms state-of-the-art methods in both Chamfer Distance and F-Score, particularly for geometrically complex structures, without requiring foreground masks.