MaskingDepth: Masked Consistency Regularization for Semi-supervised Monocular Depth Estimation

TL;DR

MaskingDepth introduces a semi-supervised framework for monocular depth estimation that leverages consistency regularization with novel masking augmentation and uncertainty estimation, reducing reliance on ground-truth depths.

Contribution

The paper presents MaskingDepth, a new semi-supervised approach with a novel masking augmentation and uncertainty-based pseudo-label selection for improved depth estimation.

Findings

Effective with fewer labeled images

Outperforms state-of-the-art semi-supervised methods

Easily extendable to domain adaptation

Abstract

We propose MaskingDepth, a novel semi-supervised learning framework for monocular depth estimation to mitigate the reliance on large ground-truth depth quantities. MaskingDepth is designed to enforce consistency between the strongly-augmented unlabeled data and the pseudo-labels derived from weakly-augmented unlabeled data, which enables learning depth without supervision. In this framework, a novel data augmentation is proposed to take the advantage of a naive masking strategy as an augmentation, while avoiding its scale ambiguity problem between depths from weakly- and strongly-augmented branches and risk of missing small-scale instances. To only retain high-confident depth predictions from the weakly-augmented branch as pseudo-labels, we also present an uncertainty estimation technique, which is used to define robust consistency regularization. Experiments on KITTI and NYU-Depth-v2 datasets demonstrate the effectiveness of each component, its robustness to the use of fewer depth-annotated images, and superior performance compared to other state-of-the-art semi-supervised methods for monocular depth estimation. Furthermore, we show our method can be easily extended to domain adaptation task. Our code is available at https://github.com/KU-CVLAB/MaskingDepth.