EvalMVX: A Unified Benchmarking for Neural 3D Reconstruction under Diverse Multiview Setups

TL;DR

EvalMVX introduces a comprehensive real-world dataset and benchmark for evaluating neural 3D reconstruction methods across diverse multiview setups, including MVS, MVPS, and MVSfP, to advance research in this field.

Contribution

The paper presents EvalMVX, a new dataset and benchmark for simultaneous quantitative assessment of multiple neural 3D reconstruction techniques under varied conditions.

Findings

Evaluated 13 recent MVX methods using EvalMVX dataset.

Identified the best-performing methods across different scenarios.

Highlighted open problems and challenges in multiview 3D reconstruction.

Abstract

Recent advancements in neural surface reconstruction have significantly enhanced 3D reconstruction. However, current real world datasets mainly focus on benchmarking multiview stereo (MVS) based on RGB inputs. Multiview photometric stereo (MVPS) and multiview shape from polarization (MVSfP), though indispensable on high-fidelity surface reconstruction and sparse inputs, have not been quantitatively assessed together with MVS. To determine the working range of different MVX (MVS, MVSfP, and MVPS) techniques, we propose EvalMVX, a real-world dataset containing $25$ objects, each captured with a polarized camera under $20$ varying views and $17$ light conditions including OLAT and natural illumination, leading to $8,500$ images. Each object includes aligned ground-truth 3D mesh, facilitating quantitative benchmarking of MVX methods simultaneously. Based on our EvalMVX, we evaluate $13$ MVX methods published in recent years, record the best-performing methods, and identify open problems under diverse geometric details and reflectance types. We hope EvalMVX and the benchmarking results can inspire future research on multiview 3D reconstruction.