Enhancing Monocular Depth Estimation with Multi-Source Auxiliary Tasks

TL;DR

This paper introduces a method that leverages auxiliary datasets from related vision tasks to improve monocular depth estimation, achieving significant quality gains and data efficiency even with limited high-quality labeled data.

Contribution

It proposes an alternating training scheme with a shared decoder that effectively incorporates auxiliary datasets, especially semantic segmentation, to enhance MDE performance.

Findings

MDE quality improves by approximately 11% with auxiliary data.

Using semantic segmentation datasets yields additional gains.

Method reduces dataset size by at least 80% while maintaining quality.

Abstract

Monocular depth estimation (MDE) is a challenging task in computer vision, often hindered by the cost and scarcity of high-quality labeled datasets. We tackle this challenge using auxiliary datasets from related vision tasks for an alternating training scheme with a shared decoder built on top of a pre-trained vision foundation model, while giving a higher weight to MDE. Through extensive experiments we demonstrate the benefits of incorporating various in-domain auxiliary datasets and tasks to improve MDE quality on average by ~11%. Our experimental analysis shows that auxiliary tasks have different impacts, confirming the importance of task selection, highlighting that quality gains are not achieved by merely adding data. Remarkably, our study reveals that using semantic segmentation datasets as Multi-Label Dense Classification (MLDC) often results in additional quality gains. Lastly, our method significantly improves the data efficiency for the considered MDE datasets, enhancing their quality while reducing their size by at least 80%. This paves the way for using auxiliary data from related tasks to improve MDE quality despite limited availability of high-quality labeled data. Code is available at https://jugit.fz-juelich.de/ias-8/mdeaux.