ErpGS: Equirectangular Image Rendering enhanced with 3D Gaussian Regularization

TL;DR

ErpGS is a novel method that improves 3D Gaussian regularization for equirectangular images, addressing distortion issues to enhance the accuracy of novel view synthesis in 360-degree imaging.

Contribution

It introduces ErpGS, a new approach that incorporates geometric and scale regularization, distortion-aware weights, and obstacle suppression to improve rendering accuracy.

Findings

ErpGS outperforms conventional methods in rendering accuracy.

The method effectively suppresses distortion effects in 360-degree images.

Experiments on public datasets validate the improvements.

Abstract

The use of multi-view images acquired by a 360-degree camera can reconstruct a 3D space with a wide area. There are 3D reconstruction methods from equirectangular images based on NeRF and 3DGS, as well as Novel View Synthesis (NVS) methods. On the other hand, it is necessary to overcome the large distortion caused by the projection model of a 360-degree camera when equirectangular images are used. In 3DGS-based methods, the large distortion of the 360-degree camera model generates extremely large 3D Gaussians, resulting in poor rendering accuracy. We propose ErpGS, which is Omnidirectional GS based on 3DGS to realize NVS addressing the problems. ErpGS introduce some rendering accuracy improvement techniques: geometric regularization, scale regularization, and distortion-aware weights and a mask to suppress the effects of obstacles in equirectangular images. Through experiments on public datasets, we demonstrate that ErpGS can render novel view images more accurately than conventional methods.