A Comparative Evaluation of Geometric Accuracy in NeRF and Gaussian Splatting

TL;DR

This paper introduces an evaluation pipeline and benchmark for assessing the geometric accuracy of neural rendering methods like NeRF and Gaussian Splatting across diverse scenes, emphasizing surface and shape fidelity.

Contribution

It provides a systematic assessment framework and benchmark focusing on geometric fidelity, addressing a gap in traditional visual quality metrics.

Findings

Benchmark includes 19 diverse scenes for evaluation.

The pipeline enables comparison of surface and shape accuracy.

Highlights the importance of geometric metrics in neural rendering evaluation.

Abstract

Recent advances in neural rendering have introduced numerous 3D scene representations. Although standard computer vision metrics evaluate the visual quality of generated images, they often overlook the fidelity of surface geometry. This limitation is particularly critical in robotics, where accurate geometry is essential for tasks such as grasping and object manipulation. In this paper, we present an evaluation pipeline for neural rendering methods that focuses on geometric accuracy, along with a benchmark comprising 19 diverse scenes. Our approach enables a systematic assessment of reconstruction methods in terms of surface and shape fidelity, complementing traditional visual metrics.