Semi-Supervised Learning with Mutual Distillation for Monocular Depth Estimation

TL;DR

This paper introduces a semi-supervised learning framework for monocular depth estimation that combines supervised and unsupervised methods through mutual distillation and diverse data augmentation, improving robustness and performance.

Contribution

It presents a novel mutual distillation approach with separate branches and tailored data augmentation, addressing limitations of existing semi-supervised methods.

Findings

Outperforms recent state-of-the-art methods

Demonstrates robustness through extensive ablation studies

Effective integration of supervised and unsupervised losses

Abstract

We propose a semi-supervised learning framework for monocular depth estimation. Compared to existing semi-supervised learning methods, which inherit limitations of both sparse supervised and unsupervised loss functions, we achieve the complementary advantages of both loss functions, by building two separate network branches for each loss and distilling each other through the mutual distillation loss function. We also present to apply different data augmentation to each branch, which improves the robustness. We conduct experiments to demonstrate the effectiveness of our framework over the latest methods and provide extensive ablation studies.