Cargando…
Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks
Predicting depth from a monocular image is an ill-posed and inherently ambiguous issue in computer vision. In this paper, we propose a pyramidal third-streamed network (PTSN) that recovers the depth information using a single given RGB image. PTSN uses pyramidal structure images, which can extract m...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2019
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6386885/ https://www.ncbi.nlm.nih.gov/pubmed/30736347 http://dx.doi.org/10.3390/s19030667 |
_version_ | 1783397445230657536 |
---|---|
author | Chen, Songnan Tang, Mengxia Kan, Jiangming |
author_facet | Chen, Songnan Tang, Mengxia Kan, Jiangming |
author_sort | Chen, Songnan |
collection | PubMed |
description | Predicting depth from a monocular image is an ill-posed and inherently ambiguous issue in computer vision. In this paper, we propose a pyramidal third-streamed network (PTSN) that recovers the depth information using a single given RGB image. PTSN uses pyramidal structure images, which can extract multiresolution features to improve the robustness of the network as the network input. The full connection layer is changed into fully convolutional layers with a new upconvolution structure, which reduces the network parameters and computational complexity. We propose a new loss function including scale-invariant, horizontal and vertical gradient loss that not only helps predict the depth values, but also clearly obtains local contours. We evaluate PTSN on the NYU Depth v2 dataset and the experimental results show that our depth predictions have better accuracy than competing methods. |
format | Online Article Text |
id | pubmed-6386885 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2019 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-63868852019-02-26 Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks Chen, Songnan Tang, Mengxia Kan, Jiangming Sensors (Basel) Article Predicting depth from a monocular image is an ill-posed and inherently ambiguous issue in computer vision. In this paper, we propose a pyramidal third-streamed network (PTSN) that recovers the depth information using a single given RGB image. PTSN uses pyramidal structure images, which can extract multiresolution features to improve the robustness of the network as the network input. The full connection layer is changed into fully convolutional layers with a new upconvolution structure, which reduces the network parameters and computational complexity. We propose a new loss function including scale-invariant, horizontal and vertical gradient loss that not only helps predict the depth values, but also clearly obtains local contours. We evaluate PTSN on the NYU Depth v2 dataset and the experimental results show that our depth predictions have better accuracy than competing methods. MDPI 2019-02-06 /pmc/articles/PMC6386885/ /pubmed/30736347 http://dx.doi.org/10.3390/s19030667 Text en © 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Chen, Songnan Tang, Mengxia Kan, Jiangming Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks |
title | Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks |
title_full | Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks |
title_fullStr | Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks |
title_full_unstemmed | Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks |
title_short | Predicting Depth from Single RGB Images with Pyramidal Three-Streamed Networks |
title_sort | predicting depth from single rgb images with pyramidal three-streamed networks |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6386885/ https://www.ncbi.nlm.nih.gov/pubmed/30736347 http://dx.doi.org/10.3390/s19030667 |
work_keys_str_mv | AT chensongnan predictingdepthfromsinglergbimageswithpyramidalthreestreamednetworks AT tangmengxia predictingdepthfromsinglergbimageswithpyramidalthreestreamednetworks AT kanjiangming predictingdepthfromsinglergbimageswithpyramidalthreestreamednetworks |