Generative adversarial network (GAN) for image synthesis
Recently, Conditional generative adversarial network (cGAN) plays an important role in image synthesis tasks and Vision Transformer (ViT) with self-attention mechanism have shown SOTA performance on computer vision field. In this report, I extent ViT to image synthesis tasks. I propose two ViT-based...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/158045 |
| _version_ | 1811680155330936832 |
|---|---|
| author | Hou, Boyu |
| author2 | Wen Bihan |
| author_facet | Wen Bihan Hou, Boyu |
| author_sort | Hou, Boyu |
| collection | NTU |
| description | Recently, Conditional generative adversarial network (cGAN) plays an important role in image synthesis tasks and Vision Transformer (ViT) with self-attention mechanism have shown SOTA performance on computer vision field. In this report, I extent ViT to image synthesis tasks. I propose two ViT-based generator architectures with upsampling and transposed convolution encoders and one ViT-based discriminator. I demonstrate that my models, named cViTGAN, are capable of image synthesis task. I perform experiments on six different benchmarks, the models achieve comparable performance to the baseline models. My work shows that we can achieve reasonable results with ViT-based models. |
| first_indexed | 2024-10-01T03:20:33Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/158045 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:20:33Z |
| publishDate | 2022 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1580452023-07-07T19:22:27Z Generative adversarial network (GAN) for image synthesis Hou, Boyu Wen Bihan School of Electrical and Electronic Engineering bihan.wen@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Recently, Conditional generative adversarial network (cGAN) plays an important role in image synthesis tasks and Vision Transformer (ViT) with self-attention mechanism have shown SOTA performance on computer vision field. In this report, I extent ViT to image synthesis tasks. I propose two ViT-based generator architectures with upsampling and transposed convolution encoders and one ViT-based discriminator. I demonstrate that my models, named cViTGAN, are capable of image synthesis task. I perform experiments on six different benchmarks, the models achieve comparable performance to the baseline models. My work shows that we can achieve reasonable results with ViT-based models. Bachelor of Engineering (Electrical and Electronic Engineering) 2022-05-26T06:49:48Z 2022-05-26T06:49:48Z 2022 Final Year Project (FYP) Hou, B. (2022). Generative adversarial network (GAN) for image synthesis. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/158045 https://hdl.handle.net/10356/158045 en A3277-211 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Hou, Boyu Generative adversarial network (GAN) for image synthesis |
| title | Generative adversarial network (GAN) for image synthesis |
| title_full | Generative adversarial network (GAN) for image synthesis |
| title_fullStr | Generative adversarial network (GAN) for image synthesis |
| title_full_unstemmed | Generative adversarial network (GAN) for image synthesis |
| title_short | Generative adversarial network (GAN) for image synthesis |
| title_sort | generative adversarial network gan for image synthesis |
| topic | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision |
| url | https://hdl.handle.net/10356/158045 |
| work_keys_str_mv | AT houboyu generativeadversarialnetworkganforimagesynthesis |