Learning Inverse Depth Regression for Multi-View Stereo with Correlation Cost Volume

TL;DR

This paper introduces a novel lightweight cost volume construction using group-wise correlation and recasts multi-view depth inference as an inverse depth regression task, achieving state-of-the-art results in scalability and accuracy.

Contribution

It proposes a new correlation-based cost volume and a cascade 3D U-Net for regularization, enabling efficient and accurate large-scale multi-view stereo depth estimation.

Findings

Achieves state-of-the-art results on DTU and Tanks and Temples datasets.

Reduces memory consumption and computational burden.

Enables sub-pixel depth estimation for large-scale scenes.

Abstract

Deep learning has shown to be effective for depth inference in multi-view stereo (MVS). However, the scalability and accuracy still remain an open problem in this domain. This can be attributed to the memory-consuming cost volume representation and inappropriate depth inference. Inspired by the group-wise correlation in stereo matching, we propose an average group-wise correlation similarity measure to construct a lightweight cost volume. This can not only reduce the memory consumption but also reduce the computational burden in the cost volume filtering. Based on our effective cost volume representation, we propose a cascade 3D U-Net module to regularize the cost volume to further boost the performance. Unlike the previous methods that treat multi-view depth inference as a depth regression problem or an inverse depth classification problem, we recast multi-view depth inference as an inverse depth regression task. This allows our network to achieve sub-pixel estimation and be applicable to large-scale scenes. Through extensive experiments on DTU dataset and Tanks and Temples dataset, we show that our proposed network with Correlation cost volume and Inverse DEpth Regression (CIDER), achieves state-of-the-art results, demonstrating its superior performance on scalability and accuracy.