EvalMVX: A Unified Benchmarking for Neural 3D Reconstruction under Diverse Multiview Setups
This addresses a gap for researchers in 3D reconstruction by providing a comprehensive benchmark, though it is incremental as it builds on existing datasets and methods.
The paper tackles the lack of unified benchmarking for neural 3D reconstruction methods under diverse multiview setups by introducing EvalMVX, a real-world dataset with 25 objects, 8,500 images, and ground-truth 3D meshes, and evaluates 13 methods to identify best performers and open problems.
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.