The University of Southampton
University of Southampton Institutional Repository

Depth estimation from a single omnidirectional image using domain adaptation

Depth estimation from a single omnidirectional image using domain adaptation
Depth estimation from a single omnidirectional image using domain adaptation
Omnidirectional cameras are becoming popular in various applications owing to their ability to capture the full surrounding scene in real-time. However, depth estimation for an omnidirectional scene is more difficult than normal perspective images due to its different system properties and distortions. It is hard to use normal depth estimation methods such as stereo matching or RGB-D sensing. A deep-learning-based single-shot depth estimation approach can be a good solution, but it requires a large labelled dataset for training. The 3D60 dataset, the largest omnidirectional dataset with depth labels, is not applicable for general scene depth estimation because it covers very limited scenes. In order to overcome this limitation, we propose a depth estimation architecture for a single omnidirectional image using domain adaptation. The proposed architecture gets labelled source domain and unlabelled target domain data together as its input and estimated depth information of the target domain using the Generative Adversarial Networks (GAN) based method. The proposed architecture shows >10% higher accuracy in depth estimation than traditional encoder-decoder models with a limited labelled dataset.
depth estimation, domain adaptation, omnidirectional image, single image
Wu, Yihong
2876bede-25f1-47a5-9e08-b98be99b2d31
Heng, Yuwen
a3edf9da-2d3b-450c-8d6d-85f76c861849
Niranjan, Mahesan
5cbaeea8-7288-4b55-a89c-c43d212ddd4f
Kim, Hansung
2c7c135c-f00b-4409-acb2-85b3a9e8225f
Wu, Yihong
2876bede-25f1-47a5-9e08-b98be99b2d31
Heng, Yuwen
a3edf9da-2d3b-450c-8d6d-85f76c861849
Niranjan, Mahesan
5cbaeea8-7288-4b55-a89c-c43d212ddd4f
Kim, Hansung
2c7c135c-f00b-4409-acb2-85b3a9e8225f

Wu, Yihong, Heng, Yuwen, Niranjan, Mahesan and Kim, Hansung (2021) Depth estimation from a single omnidirectional image using domain adaptation. 18th ACM SIGGRAPH European Conference on Visual Media Production, CVMP 2021, , Virtual, Online, United Kingdom. 06 - 07 Dec 2021. 9 pp . (doi:10.1145/3485441.3485649).

Record type: Conference or Workshop Item (Paper)

Abstract

Omnidirectional cameras are becoming popular in various applications owing to their ability to capture the full surrounding scene in real-time. However, depth estimation for an omnidirectional scene is more difficult than normal perspective images due to its different system properties and distortions. It is hard to use normal depth estimation methods such as stereo matching or RGB-D sensing. A deep-learning-based single-shot depth estimation approach can be a good solution, but it requires a large labelled dataset for training. The 3D60 dataset, the largest omnidirectional dataset with depth labels, is not applicable for general scene depth estimation because it covers very limited scenes. In order to overcome this limitation, we propose a depth estimation architecture for a single omnidirectional image using domain adaptation. The proposed architecture gets labelled source domain and unlabelled target domain data together as its input and estimated depth information of the target domain using the Generative Adversarial Networks (GAN) based method. The proposed architecture shows >10% higher accuracy in depth estimation than traditional encoder-decoder models with a limited labelled dataset.

Text
CVMP2021_camera_ready
Restricted to Repository staff only
Request a copy

More information

Accepted/In Press date: 8 September 2021
Published date: 6 December 2021
Additional Information: Funding Information: This work was supported by the EPSRC Programme Grant Immersive Audio-Visual 3D Scene Reproduction Using a Single 360 Camera (EP/V03538X/1). Publisher Copyright: © 2021 ACM. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.
Venue - Dates: 18th ACM SIGGRAPH European Conference on Visual Media Production, CVMP 2021, , Virtual, Online, United Kingdom, 2021-12-06 - 2021-12-07
Keywords: depth estimation, domain adaptation, omnidirectional image, single image

Identifiers

Local EPrints ID: 451974
URI: http://eprints.soton.ac.uk/id/eprint/451974
PURE UUID: ed90677d-b291-414a-b988-8e13efe4b4f9
ORCID for Yuwen Heng: ORCID iD orcid.org/0000-0003-3793-4811
ORCID for Mahesan Niranjan: ORCID iD orcid.org/0000-0001-7021-140X
ORCID for Hansung Kim: ORCID iD orcid.org/0000-0003-4907-0491

Catalogue record

Date deposited: 05 Nov 2021 17:30
Last modified: 28 Apr 2022 02:31

Export record

Altmetrics

Contributors

Author: Yihong Wu
Author: Yuwen Heng ORCID iD
Author: Mahesan Niranjan ORCID iD
Author: Hansung Kim ORCID iD

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×