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

We propose MaskingDepth, a semi-supervised learning framework for monocular depth estimation. MaskingDepth is designed to enforce consistency between the depths obtained from strongly-augmented images and the pseudo-depths derived from weakly-augmented images, which enables mitigating the reliance on large ground-truth depth quantities. In this framework, we leverage uncertainty estimation to only retain high-confident depth predictions from the weakly-augmented branch as pseudo-depths. We also present a novel data augmentation, dubbed K-way disjoint masking, that takes advantage of a naïve token 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 objects. 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 learning methods for monocular depth estimation.

Authors

• Jongbeom Baek
• Gyeongnyeon Kim
• Seonghoon Park 0002
• Honggyu An
• Matteo Poggi
• Seungryong Kim

Keywords

• Uncertainty
• Head
• Filtering
• Depth measurement
• Noise
• Estimation
• Semisupervised learning
• Data augmentation
• Robustness
• Intelligent robots
• Depth Estimation
• Consistency Regularization
• Monocular Depth Estimation
• Uncertainty Estimation
• Semi-supervised Learning
• Semi-supervised Methods
• Ground Truth Depth
• Masking Strategy
• Semi-supervised Learning Framework
• Scale Ambiguity
• Loss Function
• Root Mean Square Error
• Mean Square Error
• Transformer
• Decoding
• Supervised Learning
• Depth Map
• Unlabeled Data
• Self-supervised Learning
• KITTI Dataset
• Stereo Pairs
• Data Augmentation Techniques
• Pseudo Labels
• Consistency Loss
• Masking Technique
• Inherent Ambiguity
• Global Interaction
• Object Boundaries
• Self-supervised Learning Methods