Digging into Depth Priors for Outdoor Neural Radiance Fields

TL;DR

This paper thoroughly investigates how different depth priors and their usage strategies affect outdoor neural radiance fields, providing insights to improve model training in sparse viewpoint scenarios.

Contribution

It offers a comprehensive evaluation of various depth priors and usage methods for outdoor NeRFs, filling a gap in understanding their relative effectiveness.

Findings

Certain depth priors significantly improve NeRF performance.

The way depth priors are integrated impacts the quality of the results.

Recommendations for selecting and using depth priors in outdoor NeRF applications.

Abstract

Neural Radiance Fields (NeRF) have demonstrated impressive performance in vision and graphics tasks, such as novel view synthesis and immersive reality. However, the shape-radiance ambiguity of radiance fields remains a challenge, especially in the sparse viewpoints setting. Recent work resorts to integrating depth priors into outdoor NeRF training to alleviate the issue. However, the criteria for selecting depth priors and the relative merits of different priors have not been thoroughly investigated. Moreover, the relative merits of selecting different approaches to use the depth priors is also an unexplored problem. In this paper, we provide a comprehensive study and evaluation of employing depth priors to outdoor neural radiance fields, covering common depth sensing technologies and most application ways. Specifically, we conduct extensive experiments with two representative NeRF methods equipped with four commonly-used depth priors and different depth usages on two widely used outdoor datasets. Our experimental results reveal several interesting findings that can potentially benefit practitioners and researchers in training their NeRF models with depth priors. Project Page: https://cwchenwang.github.io/outdoor-nerf-depth