Non-uniform Sampling Strategies for NeRF on 360{\textdegree} images

TL;DR

This paper introduces two non-uniform sampling techniques tailored for neural radiance fields (NeRF) to improve 360-degree omnidirectional image synthesis, addressing distortions and wide viewing angles for better accuracy and efficiency.

Contribution

The study proposes distortion-aware and content-aware ray sampling schemes specifically designed for 360-degree images, enhancing NeRF performance and applicability.

Findings

Improved NeRF accuracy for 360-degree images.

Enhanced efficiency in view synthesis.

Applicable to advanced NeRF variants.

Abstract

In recent years, the performance of novel view synthesis using perspective images has dramatically improved with the advent of neural radiance fields (NeRF). This study proposes two novel techniques that effectively build NeRF for 360{\textdegree} omnidirectional images. Due to the characteristics of a 360{\textdegree} image of ERP format that has spatial distortion in their high latitude regions and a 360{\textdegree} wide viewing angle, NeRF's general ray sampling strategy is ineffective. Hence, the view synthesis accuracy of NeRF is limited and learning is not efficient. We propose two non-uniform ray sampling schemes for NeRF to suit 360{\textdegree} images - distortion-aware ray sampling and content-aware ray sampling. We created an evaluation dataset Synth360 using Replica and SceneCity models of indoor and outdoor scenes, respectively. In experiments, we show that our proposal successfully builds 360{\textdegree} image NeRF in terms of both accuracy and efficiency. The proposal is widely applicable to advanced variants of NeRF. DietNeRF, AugNeRF, and NeRF++ combined with the proposed techniques further improve the performance. Moreover, we show that our proposed method enhances the quality of real-world scenes in 360{\textdegree} images. Synth360: https://drive.google.com/drive/folders/1suL9B7DO2no21ggiIHkH3JF3OecasQLb.