A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation

In many applications, such as robotic perception, scene understanding, augmented reality, 3D reconstruction, and medical image analysis, depth from images is a fundamentally ill-posed problem. The success of depth estimation models relies on assembling a suitably large and diverse training dataset a...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Faisal Khan, Shahid Hussain, Shubhajit Basak, Mohamed Moustafa, Peter Corcoran
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Datasets deep learning depth datasets depth estimation depth loss function Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/6f4053d3f9fa4bbeb34d9231724ee45f
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:6f4053d3f9fa4bbeb34d9231724ee45f
record_format	dspace
spelling	oai:doaj.org-article:6f4053d3f9fa4bbeb34d9231724ee45f2021-11-18T00:07:22ZA Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation2169-353610.1109/ACCESS.2021.3124978https://doaj.org/article/6f4053d3f9fa4bbeb34d9231724ee45f2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9598847/https://doaj.org/toc/2169-3536In many applications, such as robotic perception, scene understanding, augmented reality, 3D reconstruction, and medical image analysis, depth from images is a fundamentally ill-posed problem. The success of depth estimation models relies on assembling a suitably large and diverse training dataset and on the selection of appropriate loss functions. It is critical for researchers in this field to be made aware of the wide range of publicly available depth datasets along with the properties of various loss functions that have been applied to depth estimation. Selection of the right training data combined with appropriate loss functions will accelerate new research and enable better comparison with state-of-the-art. Accordingly, this work offers a comprehensive review of available depth datasets as well as the loss functions that are applied in this problem domain. These depth datasets are categorised into five primary categories based on their application, namely (i) people detection and action recognition, (ii) faces and facial pose, (iii) perception-based navigation (i.e., street signs, roads), (iv) object and scene recognition, and (v) medical applications. The important characteristics and properties of each depth dataset are described and compared. A mixing strategy for depth datasets is presented in order to generalise model results across different environments and use cases. Furthermore, depth estimation loss functions that can help with training deep learning depth estimation models across different datasets are discussed. State-of-the-art deep learning-based depth estimation methods evaluations are presented for three of the most popular datasets. Finally, a discussion about challenges and future research along with recommendations for building comprehensive depth datasets will be presented as to help researchers in the selection of appropriate datasets and loss functions for evaluating their results and algorithms.Faisal KhanShahid HussainShubhajit BasakMohamed MoustafaPeter CorcoranIEEEarticleDatasetsdeep learningdepth datasetsdepth estimationdepth loss functionElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 148479-148503 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Datasets deep learning depth datasets depth estimation depth loss function Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Datasets deep learning depth datasets depth estimation depth loss function Electrical engineering. Electronics. Nuclear engineering TK1-9971 Faisal Khan Shahid Hussain Shubhajit Basak Mohamed Moustafa Peter Corcoran A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation
description	In many applications, such as robotic perception, scene understanding, augmented reality, 3D reconstruction, and medical image analysis, depth from images is a fundamentally ill-posed problem. The success of depth estimation models relies on assembling a suitably large and diverse training dataset and on the selection of appropriate loss functions. It is critical for researchers in this field to be made aware of the wide range of publicly available depth datasets along with the properties of various loss functions that have been applied to depth estimation. Selection of the right training data combined with appropriate loss functions will accelerate new research and enable better comparison with state-of-the-art. Accordingly, this work offers a comprehensive review of available depth datasets as well as the loss functions that are applied in this problem domain. These depth datasets are categorised into five primary categories based on their application, namely (i) people detection and action recognition, (ii) faces and facial pose, (iii) perception-based navigation (i.e., street signs, roads), (iv) object and scene recognition, and (v) medical applications. The important characteristics and properties of each depth dataset are described and compared. A mixing strategy for depth datasets is presented in order to generalise model results across different environments and use cases. Furthermore, depth estimation loss functions that can help with training deep learning depth estimation models across different datasets are discussed. State-of-the-art deep learning-based depth estimation methods evaluations are presented for three of the most popular datasets. Finally, a discussion about challenges and future research along with recommendations for building comprehensive depth datasets will be presented as to help researchers in the selection of appropriate datasets and loss functions for evaluating their results and algorithms.
format	article
author	Faisal Khan Shahid Hussain Shubhajit Basak Mohamed Moustafa Peter Corcoran
author_facet	Faisal Khan Shahid Hussain Shubhajit Basak Mohamed Moustafa Peter Corcoran
author_sort	Faisal Khan
title	A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation
title_short	A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation
title_full	A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation
title_fullStr	A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation
title_full_unstemmed	A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation
title_sort	review of benchmark datasets and training loss functions in neural depth estimation
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/6f4053d3f9fa4bbeb34d9231724ee45f
work_keys_str_mv	AT faisalkhan areviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT shahidhussain areviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT shubhajitbasak areviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT mohamedmoustafa areviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT petercorcoran areviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT faisalkhan reviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT shahidhussain reviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT shubhajitbasak reviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT mohamedmoustafa reviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation AT petercorcoran reviewofbenchmarkdatasetsandtraininglossfunctionsinneuraldepthestimation
_version_	1718425230192410624

A Review of Benchmark Datasets and Training Loss Functions in Neural Depth Estimation

Ejemplares similares