Digging into Depth Priors for Outdoor Neural Radiance Fields
This work addresses a key challenge for practitioners and researchers in computer vision and graphics by providing empirical insights into selecting and using depth priors for outdoor NeRF, though it is incremental as it builds on existing methods.
The paper tackles the problem of shape-radiance ambiguity in outdoor Neural Radiance Fields (NeRF) by comprehensively evaluating depth priors, finding that certain priors and usage methods improve performance, with specific gains like up to 20% reduction in error metrics on sparse-view datasets.
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