Single-shot image generation and stylisation via cross-domain correspondance
The advent of generative adversarial networks has led to many state-of-the-art methodologies in the field of image generation and stylisation. Among the most popular methods to generate new and diverse images is cross-domain correspondence, where the generated output would be a mix of the stylistic...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/163314 |
| _version_ | 1826119983459991552 |
|---|---|
| author | Kalyan, Harikishan |
| author2 | Lin Guosheng |
| author_facet | Lin Guosheng Kalyan, Harikishan |
| author_sort | Kalyan, Harikishan |
| collection | NTU |
| description | The advent of generative adversarial networks has led to many state-of-the-art methodologies in the field of image generation and stylisation. Among the most popular methods to generate new and diverse images is cross-domain correspondence, where the generated output would be a mix of the stylistic elements and attributes of a source dataset and a target dataset. This method, however, can be resource intensive due to the need for massive datasets. Existing methodologies from Ojha et al and Mind the Gap have attempted to address this issue by requiring only a few images for domain adaptation, they are prone to overfitting concerns due to a limited dataset. To counter these problems, a CLIP guided domain adaptation approach is proposed where only a single image is needed for the model to generate diverse images of various styles |
| first_indexed | 2024-10-01T05:09:05Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/163314 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:09:05Z |
| publishDate | 2022 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1633142022-12-02T00:39:12Z Single-shot image generation and stylisation via cross-domain correspondance Kalyan, Harikishan Lin Guosheng School of Computer Science and Engineering gslin@ntu.edu.sg Engineering::Computer science and engineering The advent of generative adversarial networks has led to many state-of-the-art methodologies in the field of image generation and stylisation. Among the most popular methods to generate new and diverse images is cross-domain correspondence, where the generated output would be a mix of the stylistic elements and attributes of a source dataset and a target dataset. This method, however, can be resource intensive due to the need for massive datasets. Existing methodologies from Ojha et al and Mind the Gap have attempted to address this issue by requiring only a few images for domain adaptation, they are prone to overfitting concerns due to a limited dataset. To counter these problems, a CLIP guided domain adaptation approach is proposed where only a single image is needed for the model to generate diverse images of various styles Bachelor of Engineering (Computer Engineering) 2022-12-02T00:39:12Z 2022-12-02T00:39:12Z 2022 Final Year Project (FYP) Kalyan, H. (2022). Single-shot image generation and stylisation via cross-domain correspondance. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/163314 https://hdl.handle.net/10356/163314 en application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering Kalyan, Harikishan Single-shot image generation and stylisation via cross-domain correspondance |
| title | Single-shot image generation and stylisation via cross-domain correspondance |
| title_full | Single-shot image generation and stylisation via cross-domain correspondance |
| title_fullStr | Single-shot image generation and stylisation via cross-domain correspondance |
| title_full_unstemmed | Single-shot image generation and stylisation via cross-domain correspondance |
| title_short | Single-shot image generation and stylisation via cross-domain correspondance |
| title_sort | single shot image generation and stylisation via cross domain correspondance |
| topic | Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/163314 |
| work_keys_str_mv | AT kalyanharikishan singleshotimagegenerationandstylisationviacrossdomaincorrespondance |