ErpGS: Equirectangular Image Rendering enhanced with 3D Gaussian Regularization

2025-05-26

AI Summary

Equirectangular projections (ERP) from 360° cameras suffer severe geometric distortion, causing excessive elongation of Gaussian ellipsoids and degraded rendering fidelity in 3D Gaussian Splatting (3DGS). Method: We propose the first geometry- and scale-aware regularization framework tailored for omnidirectional imaging. Our approach jointly incorporates spherical geometric constraints, adaptive scale regularization, distortion-aware weighted rendering, and occlusion masks informed by scene obstacles. This enables end-to-end adaptation of 3DGS to ERP inputs. Contribution/Results: Evaluated on public 360° datasets, our method achieves a PSNR gain of over 2.1 dB compared to standard 3DGS. Structural fidelity is significantly improved near poles and image boundaries. The framework establishes a robust, efficient paradigm for high-fidelity novel-view synthesis from omnidirectional imagery.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

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.

Problem

• Overcome distortion in 360-degree equirectangular image projection
• Improve rendering accuracy in 3D Gaussian-based methods
• Suppress obstacle effects in equirectangular images

Innovation

• 3D Gaussian regularization for equirectangular images
• Geometric and scale regularization techniques
• Distortion-aware weights and obstacle suppression