Non-uniform Sampling Strategies for NeRF on 360{\textdegree} images
This work addresses a domain-specific challenge in novel view synthesis for 360° images, offering incremental improvements to existing NeRF methods.
The paper tackled the problem of applying neural radiance fields (NeRF) to 360° omnidirectional images, which suffer from spatial distortion and wide viewing angles, by proposing two non-uniform ray sampling schemes (distortion-aware and content-aware) to improve accuracy and efficiency. The result showed successful application to both synthetic and real-world scenes, with enhanced performance when combined with advanced NeRF variants like DietNeRF, AugNeRF, and NeRF++.
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.