HistReNeRF: Historic Image Relocalisation within Contemporary Neural Radiance Field Reconstructions

2026-08-15

AI Summary

This study addresses the challenge of relocalizing historical photographs, which is complicated by significant disparities in appearance and layout. To overcome this, we propose HistReNeRF, a framework that leverages Neural Radiance Fields (NeRF) as a continuous query interface. By matching DINOv2 features with sampled rays directly within the feature space, the method achieves cross-temporal domain adaptation and estimates 6-DoF poses without modifying query images. Experimental evaluations on cross-temporal datasets demonstrate that HistReNeRF reduces translation and rotation errors by 11% and 16%, respectively. These results effectively validate the superiority and robustness of NeRF-based approaches for cross-domain scene relocalization in challenging historical contexts.

Abstract

Relocalising archival photographs within a contemporary scene model is challenging because historic and modern views can differ in photographic appearance, visible objects, and spatial layout. Therefore, we present HistReNeRF, a framework that estimates the 6-DoF pose of a historic photograph by matching adapted DINOv2 patch features to candidate rays sampled from a contemporary Neural Radiance Field (NeRF) reconstruction. The continuous representation of a NeRF provides a queryable scene interface from which candidate rays can be sampled and matched, enabling domain adaptation between historic photography and contemporary images directly in the feature representation used for localisation. We evaluate embedding-space-based domain adaptation against pixel-space methods on a new cross-temporal dataset comprising 10,545 contemporary street-level images and 230 archival photographs from three European landmarks. Embedding-space adaptation reduces translation and rotation errors by an average of 11% and 16%, respectively, across the three scenes. These results show that neural scene relocalisation provides a natural interface for feature-space adaptation, reducing cross-temporal appearance shift without modifying the query image. Code and dataset at https://github.com/ARTUROLab/HistReNeRF.

Problem

• Historic Image Relocalisation
• Cross-temporal Domain Adaptation
• Neural Radiance Fields
• 6-DoF Pose Estimation

Innovation

• HistReNeRF
• Neural Radiance Field
• Cross-temporal Relocalisation
• Feature-space Domain Adaptation
• DINOv2