Omnidirectional depth estimation for semantic segmentation
Omnidirectional depth estimation for semantic segmentation
This research presents a comprehensive system encompassing semantic segmentation and depth estimation for 360-degree images. It introduces effective methodologies to tackle the challenges associated with depth estimation in panoramic imagery and enhance the precision of semantic segmentation. This article is primarily divided into two sections. The first section emphasizes the significance of integrating depth information in semantic segmentation tasks by comparing its impact to cases where it is not utilized. The second part delves into the discussion of three different approaches to address the spherical distortion in 360-degree images and constructs neural networks for depth estimation. A variety of evaluation metrics are employed to analyze, assess, and compare the results of these three methods while exploring their respective advantages and drawbacks.
360-degree images, Deeping learning, Depth estimation, Semantic segmentation
Zhou, Jiaqi
d44f8a96-b2f4-4c67-9674-b8f121a169aa
Wu, Yihong
2876bede-25f1-47a5-9e08-b98be99b2d31
Hardy, Peter Timothy David
361a5d48-51cf-4eaf-9b60-1de78f2f2f20
Kim, Hansung
2c7c135c-f00b-4409-acb2-85b3a9e8225f
19 March 2024
Zhou, Jiaqi
d44f8a96-b2f4-4c67-9674-b8f121a169aa
Wu, Yihong
2876bede-25f1-47a5-9e08-b98be99b2d31
Hardy, Peter Timothy David
361a5d48-51cf-4eaf-9b60-1de78f2f2f20
Kim, Hansung
2c7c135c-f00b-4409-acb2-85b3a9e8225f
Zhou, Jiaqi, Wu, Yihong, Hardy, Peter Timothy David and Kim, Hansung
(2024)
Omnidirectional depth estimation for semantic segmentation.
In 2024 International Conference on Electronics, Information, and Communication, ICEIC 2024.
IEEE.
4 pp
.
(doi:10.1109/ICEIC61013.2024.10457236).
Record type:
Conference or Workshop Item
(Paper)
Abstract
This research presents a comprehensive system encompassing semantic segmentation and depth estimation for 360-degree images. It introduces effective methodologies to tackle the challenges associated with depth estimation in panoramic imagery and enhance the precision of semantic segmentation. This article is primarily divided into two sections. The first section emphasizes the significance of integrating depth information in semantic segmentation tasks by comparing its impact to cases where it is not utilized. The second part delves into the discussion of three different approaches to address the spherical distortion in 360-degree images and constructs neural networks for depth estimation. A variety of evaluation metrics are employed to analyze, assess, and compare the results of these three methods while exploring their respective advantages and drawbacks.
Text
130_24ICEIC-Jiaqui-Published
Restricted to Repository staff only
Request a copy
More information
Published date: 19 March 2024
Additional Information:
Publisher Copyright:
© 2024 IEEE.
Venue - Dates:
International Conference on Electronics, Information, and Communication, Taipei Marriott Hotel, Taipei, Taiwan, 2024-01-28 - 2024-01-31
Keywords:
360-degree images, Deeping learning, Depth estimation, Semantic segmentation
Identifiers
Local EPrints ID: 490619
URI: http://eprints.soton.ac.uk/id/eprint/490619
PURE UUID: 60444191-f5d3-46da-b385-59890aa06f54
Catalogue record
Date deposited: 31 May 2024 16:43
Last modified: 25 Jul 2024 02:00
Export record
Altmetrics
Contributors
Author:
Jiaqi Zhou
Author:
Yihong Wu
Author:
Peter Timothy David Hardy
Author:
Hansung Kim
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