Scale variance minimization for unsupervised domain adaptation in image segmentation
We focus on unsupervised domain adaptation (UDA) in image segmentation. Existing works address this challenge largely by aligning inter-domain representations, which may lead over-alignment that impairs the semantic structures of images and further target-domain segmentation performance. We design a...
Saved in:
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2022
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/157050 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-157050 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1570502022-05-01T06:34:41Z Scale variance minimization for unsupervised domain adaptation in image segmentation Guan, Dayan Huang, Jiaxing Lu, Shijian Xiao, Aoran School of Computer Science and Engineering Singtel Cognitive and Artificial Intelligence Lab for Enterprises (SCALE@NTU) Engineering::Computer science and engineering Unsupervised Domain Adaptation Image Segmentation We focus on unsupervised domain adaptation (UDA) in image segmentation. Existing works address this challenge largely by aligning inter-domain representations, which may lead over-alignment that impairs the semantic structures of images and further target-domain segmentation performance. We design a scale variance minimization (SVMin) method by enforcing the intra-image semantic structure consistency in the target domain. Specifically, SVMin leverages an intrinsic property that simple scale transformation has little effect on the semantic structures of images. It thus introduces certain supervision in the target domain by imposing a scale-invariance constraint while learning to segment an image and its scale-transformation concurrently. Additionally, SVMin is complementary to most existing UDA techniques and can be easily incorporated with consistent performance boost but little extra parameters. Extensive experiments show that our method achieves superior domain adaptive segmentation performance as compared with the state-of-the-art. Preliminary studies show that SVMin can be easily adapted for UDA-based image classification. Submitted/Accepted version This research was conducted at Singtel Cognitive and Artificial Intelligence Lab for Enterprises (SCALE@NTU), which is a collaboration between Singapore Telecommunications Limited (Singtel) and Nanyang Technological University (NTU) that is funded by the Singapore Government through the Industry Alignment Fund - Industry Collaboration Projects Grant. 2022-05-01T06:34:41Z 2022-05-01T06:34:41Z 2021 Journal Article Guan, D., Huang, J., Lu, S. & Xiao, A. (2021). Scale variance minimization for unsupervised domain adaptation in image segmentation. Pattern Recognition, 112, 107764-. https://dx.doi.org/10.1016/j.patcog.2020.107764 0031-3203 https://hdl.handle.net/10356/157050 10.1016/j.patcog.2020.107764 2-s2.0-85097709318 112 107764 en Pattern Recognition © 2020 Elsevier Ltd. All rights reserved. This paper was published in Pattern Recognition and is made available with permission of Elsevier Ltd. application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering Unsupervised Domain Adaptation Image Segmentation |
spellingShingle |
Engineering::Computer science and engineering Unsupervised Domain Adaptation Image Segmentation Guan, Dayan Huang, Jiaxing Lu, Shijian Xiao, Aoran Scale variance minimization for unsupervised domain adaptation in image segmentation |
description |
We focus on unsupervised domain adaptation (UDA) in image segmentation. Existing works address this challenge largely by aligning inter-domain representations, which may lead over-alignment that impairs the semantic structures of images and further target-domain segmentation performance. We design a scale variance minimization (SVMin) method by enforcing the intra-image semantic structure consistency in the target domain. Specifically, SVMin leverages an intrinsic property that simple scale transformation has little effect on the semantic structures of images. It thus introduces certain supervision in the target domain by imposing a scale-invariance constraint while learning to segment an image and its scale-transformation concurrently. Additionally, SVMin is complementary to most existing UDA techniques and can be easily incorporated with consistent performance boost but little extra parameters. Extensive experiments show that our method achieves superior domain adaptive segmentation performance as compared with the state-of-the-art. Preliminary studies show that SVMin can be easily adapted for UDA-based image classification. |
author2 |
School of Computer Science and Engineering |
author_facet |
School of Computer Science and Engineering Guan, Dayan Huang, Jiaxing Lu, Shijian Xiao, Aoran |
format |
Article |
author |
Guan, Dayan Huang, Jiaxing Lu, Shijian Xiao, Aoran |
author_sort |
Guan, Dayan |
title |
Scale variance minimization for unsupervised domain adaptation in image segmentation |
title_short |
Scale variance minimization for unsupervised domain adaptation in image segmentation |
title_full |
Scale variance minimization for unsupervised domain adaptation in image segmentation |
title_fullStr |
Scale variance minimization for unsupervised domain adaptation in image segmentation |
title_full_unstemmed |
Scale variance minimization for unsupervised domain adaptation in image segmentation |
title_sort |
scale variance minimization for unsupervised domain adaptation in image segmentation |
publishDate |
2022 |
url |
https://hdl.handle.net/10356/157050 |
_version_ |
1734310231577460736 |