Deep image enhancement
Deep-learning based methods have brought a huge improvement in the field of image restoration and enhancement. Recent methods explore generative priors from pre-trained generator such as StyleGAN for the task of restoration. In this work, I follow this direction and delve deeper to gain more insight...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/153249 |
| _version_ | 1826125774767259648 |
|---|---|
| author | Han, Jun |
| author2 | Chen Change Loy |
| author_facet | Chen Change Loy Han, Jun |
| author_sort | Han, Jun |
| collection | NTU |
| description | Deep-learning based methods have brought a huge improvement in the field of image restoration and enhancement. Recent methods explore generative priors from pre-trained generator such as StyleGAN for the task of restoration. In this work, I follow this direction and delve deeper to gain more insights. I first conduct experiments and analysis on a relatively mature task – image denoising. My experiments demonstrate that the generative priors encapsulated in a generative network (StyleGAN) is able to improve the performance in not only super-resolution but also denoising. Furthermore, I analyze the sensitivity of such networks toward the changes of the input image. I find that even a subtle change in the input could lead to substantial changes in the output. Motivated by my findings, I shift the focus to the task of real-world face image restoration, and I devise a simple yet effective image manipulation method that could largely improve the performance of the outputs of a pre-trained model. |
| first_indexed | 2024-10-01T06:41:52Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/153249 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:41:52Z |
| publishDate | 2021 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1532492021-11-17T00:57:40Z Deep image enhancement Han, Jun Chen Change Loy School of Computer Science and Engineering ccloy@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Deep-learning based methods have brought a huge improvement in the field of image restoration and enhancement. Recent methods explore generative priors from pre-trained generator such as StyleGAN for the task of restoration. In this work, I follow this direction and delve deeper to gain more insights. I first conduct experiments and analysis on a relatively mature task – image denoising. My experiments demonstrate that the generative priors encapsulated in a generative network (StyleGAN) is able to improve the performance in not only super-resolution but also denoising. Furthermore, I analyze the sensitivity of such networks toward the changes of the input image. I find that even a subtle change in the input could lead to substantial changes in the output. Motivated by my findings, I shift the focus to the task of real-world face image restoration, and I devise a simple yet effective image manipulation method that could largely improve the performance of the outputs of a pre-trained model. Bachelor of Science in Data Science and Artificial Intelligence 2021-11-17T00:57:40Z 2021-11-17T00:57:40Z 2021 Final Year Project (FYP) Han, J. (2021). Deep image enhancement. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/153249 https://hdl.handle.net/10356/153249 en SCSE20-0824 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Han, Jun Deep image enhancement |
| title | Deep image enhancement |
| title_full | Deep image enhancement |
| title_fullStr | Deep image enhancement |
| title_full_unstemmed | Deep image enhancement |
| title_short | Deep image enhancement |
| title_sort | deep image enhancement |
| topic | Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision |
| url | https://hdl.handle.net/10356/153249 |
| work_keys_str_mv | AT hanjun deepimageenhancement |