Depth estimation for a single omnidirectional image with reversed-gradient warming-up thresholds discriminator
Depth estimation for a single omnidirectional image with reversed-gradient warming-up thresholds discriminator
Depth estimation for single image using deep learning requires a large labelled depth dataset with various scenes for training. However, currently published omnidirectional depth datasets cover limited types of scenes and are not suitable for depth estimation for various real-world scenes. With the challenge of labelled real-world datasets generation and stability of the performance, we propose an architecture with the Reverse-gradient Warming-up Threshold Discriminator (RWTD) to estimate real-world depth maps from the synthetic ground truth. It takes labelled synthetic scenes of a source domain and unlabelled real-world scenes of a target domain as inputs to predict the corresponding depth maps. Compared with state-of-the-art encoder-decoder models, the proposed architecture shows an 11% points improvement on the testing dataset for depth accuracy.
Depth estimation, domain adaptation
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
5 May 2023
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
(2023)
Depth estimation for a single omnidirectional image with reversed-gradient warming-up thresholds discriminator.
In ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, Proceedings.
vol. 2023-June,
IEEE.
5 pp
.
(doi:10.1109/ICASSP49357.2023.10094996).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Depth estimation for single image using deep learning requires a large labelled depth dataset with various scenes for training. However, currently published omnidirectional depth datasets cover limited types of scenes and are not suitable for depth estimation for various real-world scenes. With the challenge of labelled real-world datasets generation and stability of the performance, we propose an architecture with the Reverse-gradient Warming-up Threshold Discriminator (RWTD) to estimate real-world depth maps from the synthetic ground truth. It takes labelled synthetic scenes of a source domain and unlabelled real-world scenes of a target domain as inputs to predict the corresponding depth maps. Compared with state-of-the-art encoder-decoder models, the proposed architecture shows an 11% points improvement on the testing dataset for depth accuracy.
This record has no associated files available for download.
More information
Published date: 5 May 2023
Additional Information:
Funding Information:
This work was partially supported by the EPSRC Programme Grant Immersive Audio-Visual 3D Scene Reproduction (EP/V03538X/1) and partially by the Korea Institute of Science and Technology (KIST) Institutional Program (Project No. 2E31591)
Publisher Copyright:
© 2023 IEEE.
Venue - Dates:
ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), , Rhodes, Greece, 2023-06-04 - 2023-06-10
Keywords:
Depth estimation, domain adaptation
Identifiers
Local EPrints ID: 479877
URI: http://eprints.soton.ac.uk/id/eprint/479877
ISSN: 1520-6149
PURE UUID: 2bd49485-2372-40e2-aa15-f904764bc312
Catalogue record
Date deposited: 28 Jul 2023 16:34
Last modified: 18 Mar 2024 03:07
Export record
Altmetrics
Contributors
Author:
Yihong Wu
Author:
Yuwen Heng
Author:
Mahesan Niranjan
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