Robust Monocular Depth Estimation under Challenging Conditions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gasperini, Stefano, Morbitzer, Nils, Jung, HyunJun, Navab, Nassir, Tombari, Federico
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911761494441984
author Gasperini, Stefano
Morbitzer, Nils
Jung, HyunJun
Navab, Nassir
Tombari, Federico
author_facet Gasperini, Stefano
Morbitzer, Nils
Jung, HyunJun
Navab, Nassir
Tombari, Federico
contents While state-of-the-art monocular depth estimation approaches achieve impressive results in ideal settings, they are highly unreliable under challenging illumination and weather conditions, such as at nighttime or in the presence of rain. In this paper, we uncover these safety-critical issues and tackle them with md4all: a simple and effective solution that works reliably under both adverse and ideal conditions, as well as for different types of learning supervision. We achieve this by exploiting the efficacy of existing methods under perfect settings. Therefore, we provide valid training signals independently of what is in the input. First, we generate a set of complex samples corresponding to the normal training ones. Then, we train the model by guiding its self- or full-supervision by feeding the generated samples and computing the standard losses on the corresponding original images. Doing so enables a single model to recover information across diverse conditions without modifications at inference time. Extensive experiments on two challenging public datasets, namely nuScenes and Oxford RobotCar, demonstrate the effectiveness of our techniques, outperforming prior works by a large margin in both standard and challenging conditions. Source code and data are available at: https://md4all.github.io.
format Preprint
id arxiv_https___arxiv_org_abs_2308_09711
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Robust Monocular Depth Estimation under Challenging Conditions
Gasperini, Stefano
Morbitzer, Nils
Jung, HyunJun
Navab, Nassir
Tombari, Federico
Computer Vision and Pattern Recognition
Machine Learning
Robotics
While state-of-the-art monocular depth estimation approaches achieve impressive results in ideal settings, they are highly unreliable under challenging illumination and weather conditions, such as at nighttime or in the presence of rain. In this paper, we uncover these safety-critical issues and tackle them with md4all: a simple and effective solution that works reliably under both adverse and ideal conditions, as well as for different types of learning supervision. We achieve this by exploiting the efficacy of existing methods under perfect settings. Therefore, we provide valid training signals independently of what is in the input. First, we generate a set of complex samples corresponding to the normal training ones. Then, we train the model by guiding its self- or full-supervision by feeding the generated samples and computing the standard losses on the corresponding original images. Doing so enables a single model to recover information across diverse conditions without modifications at inference time. Extensive experiments on two challenging public datasets, namely nuScenes and Oxford RobotCar, demonstrate the effectiveness of our techniques, outperforming prior works by a large margin in both standard and challenging conditions. Source code and data are available at: https://md4all.github.io.
title Robust Monocular Depth Estimation under Challenging Conditions
topic Computer Vision and Pattern Recognition
Machine Learning
Robotics
url https://arxiv.org/abs/2308.09711