DAGAN: A Domain-Aware Method for Image-to-Image Translations
The image-to-image translation method aims to learn inter-domain mappings from paired/unpaired data. Although this technique has been widely used for visual predication tasks—such as classification and image segmentation—and achieved great results, we still failed to perform flexible translations wh...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2020/9341907 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832552102573899776 |
|---|---|
| author | Xu Yin Yan Li Byeong-Seok Shin |
| author_facet | Xu Yin Yan Li Byeong-Seok Shin |
| author_sort | Xu Yin |
| collection | DOAJ |
| description | The image-to-image translation method aims to learn inter-domain mappings from paired/unpaired data. Although this technique has been widely used for visual predication tasks—such as classification and image segmentation—and achieved great results, we still failed to perform flexible translations when attempting to learn different mappings, especially for images containing multiple instances. To tackle this problem, we propose a generative framework DAGAN (Domain-aware Generative Adversarial etwork) that enables domains to learn diverse mapping relationships. We assumed that an image is composed with background and instance domain and then fed them into different translation networks. Lastly, we integrated the translated domains into a complete image with smoothed labels to maintain realism. We examined the instance-aware framework on datasets generated by YOLO and confirmed that this is capable of generating images of equal or better diversity compared to current translation models. |
| format | Article |
| id | doaj-art-e3982e6b23d04bc99ca08bb61f4552c9 |
| institution | Kabale University |
| issn | 1076-2787 1099-0526 |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-e3982e6b23d04bc99ca08bb61f4552c92025-02-03T05:59:35ZengWileyComplexity1076-27871099-05262020-01-01202010.1155/2020/93419079341907DAGAN: A Domain-Aware Method for Image-to-Image TranslationsXu Yin0Yan Li1Byeong-Seok Shin2Department of Computer Engineering, Inha University, Incheon 082, Republic of KoreaDepartment of Computer Engineering, Inha University, Incheon 082, Republic of KoreaDepartment of Computer Engineering, Inha University, Incheon 082, Republic of KoreaThe image-to-image translation method aims to learn inter-domain mappings from paired/unpaired data. Although this technique has been widely used for visual predication tasks—such as classification and image segmentation—and achieved great results, we still failed to perform flexible translations when attempting to learn different mappings, especially for images containing multiple instances. To tackle this problem, we propose a generative framework DAGAN (Domain-aware Generative Adversarial etwork) that enables domains to learn diverse mapping relationships. We assumed that an image is composed with background and instance domain and then fed them into different translation networks. Lastly, we integrated the translated domains into a complete image with smoothed labels to maintain realism. We examined the instance-aware framework on datasets generated by YOLO and confirmed that this is capable of generating images of equal or better diversity compared to current translation models.http://dx.doi.org/10.1155/2020/9341907 |
| spellingShingle | Xu Yin Yan Li Byeong-Seok Shin DAGAN: A Domain-Aware Method for Image-to-Image Translations Complexity |
| title | DAGAN: A Domain-Aware Method for Image-to-Image Translations |
| title_full | DAGAN: A Domain-Aware Method for Image-to-Image Translations |
| title_fullStr | DAGAN: A Domain-Aware Method for Image-to-Image Translations |
| title_full_unstemmed | DAGAN: A Domain-Aware Method for Image-to-Image Translations |
| title_short | DAGAN: A Domain-Aware Method for Image-to-Image Translations |
| title_sort | dagan a domain aware method for image to image translations |
| url | http://dx.doi.org/10.1155/2020/9341907 |
| work_keys_str_mv | AT xuyin daganadomainawaremethodforimagetoimagetranslations AT yanli daganadomainawaremethodforimagetoimagetranslations AT byeongseokshin daganadomainawaremethodforimagetoimagetranslations |