PCM-NeRF: Probabilistic Camera Modeling for Neural Radiance Fields under Pose Uncertainty
For neural surface reconstruction, this work addresses the practical problem of inaccurate camera poses from SfM, improving robustness without requiring foreground masks.
PCM-NeRF introduces a probabilistic camera modeling framework that learns per-camera pose uncertainty to handle imperfect pose estimates, outperforming state-of-the-art methods in Chamfer Distance and F-Score on scenes with severe pose outliers.
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.