Multi-View Guided Multi-View Stereo

TL;DR

This paper presents a deep multi-view stereo framework that uses sparse depth hints and multi-view guidance to significantly improve dense 3D reconstruction accuracy from multiple images.

Contribution

It introduces a novel multi-view guidance strategy that leverages sparse depth measurements to enhance neural network-based 3D reconstruction.

Findings

Improves depth map accuracy across multiple deep MVS networks.

Enhances density and precision of 3D reconstructions.

Proves effective on BlendedMVG and DTU datasets.

Abstract

This paper introduces a novel deep framework for dense 3D reconstruction from multiple image frames, leveraging a sparse set of depth measurements gathered jointly with image acquisition. Given a deep multi-view stereo network, our framework uses sparse depth hints to guide the neural network by modulating the plane-sweep cost volume built during the forward step, enabling us to infer constantly much more accurate depth maps. Moreover, since multiple viewpoints can provide additional depth measurements, we propose a multi-view guidance strategy that increases the density of the sparse points used to guide the network, thus leading to even more accurate results. We evaluate our Multi-View Guided framework within a variety of state-of-the-art deep multi-view stereo networks, demonstrating its effectiveness at improving the results achieved by each of them on BlendedMVG and DTU datasets.