MVSBoost: An Efficient Point Cloud-based 3D Reconstruction

TL;DR

This paper introduces MVSBoost, an efficient multi-view stereo framework that combines multi-view imagery, robust camera pose estimation, and advanced processing to produce highly accurate and detailed 3D reconstructions with improved efficiency and robustness.

Contribution

The paper presents a novel MVS framework that integrates multi-view imagery with enhanced processing techniques, outperforming traditional methods and competing with neural implicit field approaches.

Findings

Achieves superior accuracy measured by Chamfer distance on synthetic datasets.

Demonstrates improved computational efficiency and robustness in complex scenes.

Produces detailed and clear 3D reconstructions capable of handling occlusions and multiple viewpoints.

Abstract

Efficient and accurate 3D reconstruction is crucial for various applications, including augmented and virtual reality, medical imaging, and cinematic special effects. While traditional Multi-View Stereo (MVS) systems have been fundamental in these applications, using neural implicit fields in implicit 3D scene modeling has introduced new possibilities for handling complex topologies and continuous surfaces. However, neural implicit fields often suffer from computational inefficiencies, overfitting, and heavy reliance on data quality, limiting their practical use. This paper presents an enhanced MVS framework that integrates multi-view 360-degree imagery with robust camera pose estimation via Structure from Motion (SfM) and advanced image processing for point cloud densification, mesh reconstruction, and texturing. Our approach significantly improves upon traditional MVS methods, offering superior accuracy and precision as validated using Chamfer distance metrics on the Realistic Synthetic 360 dataset. The developed MVS technique enhances the detail and clarity of 3D reconstructions and demonstrates superior computational efficiency and robustness in complex scene reconstruction, effectively handling occlusions and varying viewpoints. These improvements suggest that our MVS framework can compete with and potentially exceed current state-of-the-art neural implicit field methods, especially in scenarios requiring real-time processing and scalability.