An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion
Image-to-image conversion based on deep learning techniques is a topic of interest in the fields of robotics and computer vision. A series of typical tasks, such as applying semantic labels to building photos, edges to photos, and raining to de-raining, can be seen as paired image-to-image conversio...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/15/4161 |
| _version_ | 1797561197047840768 |
|---|---|
| author | Aamir Khan Weidong Jin Muqeet Ahmad Rizwan Ali Naqvi Desheng Wang |
| author_facet | Aamir Khan Weidong Jin Muqeet Ahmad Rizwan Ali Naqvi Desheng Wang |
| author_sort | Aamir Khan |
| collection | DOAJ |
| description | Image-to-image conversion based on deep learning techniques is a topic of interest in the fields of robotics and computer vision. A series of typical tasks, such as applying semantic labels to building photos, edges to photos, and raining to de-raining, can be seen as paired image-to-image conversion problems. In such problems, the image generation network learns from the information in the form of input images. The input images and the corresponding targeted images must share the same basic structure to perfectly generate target-oriented output images. However, the shared basic structure between paired images is not as ideal as assumed, which can significantly affect the output of the generating model. Therefore, we propose a novel Input-Perceptual and Reconstruction Adversarial Network (IP-RAN) as an all-purpose framework for imperfect paired image-to-image conversion problems. We demonstrate, through the experimental results, that our IP-RAN method significantly outperforms the current state-of-the-art techniques. |
| first_indexed | 2024-03-10T18:11:40Z |
| format | Article |
| id | doaj.art-02d2ac96b0154aac8bec3310af7428df |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T18:11:40Z |
| publishDate | 2020-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-02d2ac96b0154aac8bec3310af7428df2023-11-20T08:02:06ZengMDPI AGSensors1424-82202020-07-012015416110.3390/s20154161An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image ConversionAamir Khan0Weidong Jin1Muqeet Ahmad2Rizwan Ali Naqvi3Desheng Wang4School of Electrical Engineering, Southwest Jiaotong University, Chengdu 611756, ChinaSchool of Electrical Engineering, Southwest Jiaotong University, Chengdu 611756, ChinaSchool of Information Science and Technology, Southwest Jiaotong University, Chengdu 611756, ChinaDepartment of Unmanned Vehicle Engineering, Sejong University, Seoul 05006, KoreaSchool of Electrical Engineering, Southwest Jiaotong University, Chengdu 611756, ChinaImage-to-image conversion based on deep learning techniques is a topic of interest in the fields of robotics and computer vision. A series of typical tasks, such as applying semantic labels to building photos, edges to photos, and raining to de-raining, can be seen as paired image-to-image conversion problems. In such problems, the image generation network learns from the information in the form of input images. The input images and the corresponding targeted images must share the same basic structure to perfectly generate target-oriented output images. However, the shared basic structure between paired images is not as ideal as assumed, which can significantly affect the output of the generating model. Therefore, we propose a novel Input-Perceptual and Reconstruction Adversarial Network (IP-RAN) as an all-purpose framework for imperfect paired image-to-image conversion problems. We demonstrate, through the experimental results, that our IP-RAN method significantly outperforms the current state-of-the-art techniques.https://www.mdpi.com/1424-8220/20/15/4161image-to-image conversionimage de-raininglabel to photosedges to photosgenerative adversarial network (GAN) |
| spellingShingle | Aamir Khan Weidong Jin Muqeet Ahmad Rizwan Ali Naqvi Desheng Wang An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion Sensors image-to-image conversion image de-raining label to photos edges to photos generative adversarial network (GAN) |
| title | An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion |
| title_full | An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion |
| title_fullStr | An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion |
| title_full_unstemmed | An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion |
| title_short | An Input-Perceptual Reconstruction Adversarial Network for Paired Image-to-Image Conversion |
| title_sort | input perceptual reconstruction adversarial network for paired image to image conversion |
| topic | image-to-image conversion image de-raining label to photos edges to photos generative adversarial network (GAN) |
| url | https://www.mdpi.com/1424-8220/20/15/4161 |
| work_keys_str_mv | AT aamirkhan aninputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT weidongjin aninputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT muqeetahmad aninputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT rizwanalinaqvi aninputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT deshengwang aninputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT aamirkhan inputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT weidongjin inputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT muqeetahmad inputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT rizwanalinaqvi inputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion AT deshengwang inputperceptualreconstructionadversarialnetworkforpairedimagetoimageconversion |