Deep learning-based image forgery detection
In recent years, different kinds of image synthesis models have been developed, which are widely used in the field of deep learning and daily life. However, many people are troubled by these techniques because they can be utilized by malicious people to produce fake images. Nevertheless, different i...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/158878 |
| _version_ | 1826117012187774976 |
|---|---|
| author | Lu, Ye |
| author2 | Alex Chichung Kot |
| author_facet | Alex Chichung Kot Lu, Ye |
| author_sort | Lu, Ye |
| collection | NTU |
| description | In recent years, different kinds of image synthesis models have been developed, which are widely used in the field of deep learning and daily life. However, many people are troubled by these techniques because they can be utilized by malicious people to produce fake images. Nevertheless, different image synthesis techniques emerge one after another, and few existing image discrimination models can detect these different types of images. This dissertation discusses a model that can detect different image synthesis techniques, which has good practical application and research value. In addition, we introduce a brand new dataset, TMCface. We use this dataset and other public datasets to compare the performance of baselines with ours.
Keywords: Image forensics, Generalization, DeepLearning, ResNet50, TMCface. |
| first_indexed | 2024-10-01T04:20:48Z |
| format | Thesis-Master by Coursework |
| id | ntu-10356/158878 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:20:48Z |
| publishDate | 2022 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1588782023-07-04T17:48:55Z Deep learning-based image forgery detection Lu, Ye Alex Chichung Kot School of Electrical and Electronic Engineering EACKOT@ntu.edu.sg Engineering::Electrical and electronic engineering In recent years, different kinds of image synthesis models have been developed, which are widely used in the field of deep learning and daily life. However, many people are troubled by these techniques because they can be utilized by malicious people to produce fake images. Nevertheless, different image synthesis techniques emerge one after another, and few existing image discrimination models can detect these different types of images. This dissertation discusses a model that can detect different image synthesis techniques, which has good practical application and research value. In addition, we introduce a brand new dataset, TMCface. We use this dataset and other public datasets to compare the performance of baselines with ours. Keywords: Image forensics, Generalization, DeepLearning, ResNet50, TMCface. Master of Science (Signal Processing) 2022-05-31T05:53:38Z 2022-05-31T05:53:38Z 2022 Thesis-Master by Coursework Lu, Y. (2022). Deep learning-based image forgery detection. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/158878 https://hdl.handle.net/10356/158878 en application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Electrical and electronic engineering Lu, Ye Deep learning-based image forgery detection |
| title | Deep learning-based image forgery detection |
| title_full | Deep learning-based image forgery detection |
| title_fullStr | Deep learning-based image forgery detection |
| title_full_unstemmed | Deep learning-based image forgery detection |
| title_short | Deep learning-based image forgery detection |
| title_sort | deep learning based image forgery detection |
| topic | Engineering::Electrical and electronic engineering |
| url | https://hdl.handle.net/10356/158878 |
| work_keys_str_mv | AT luye deeplearningbasedimageforgerydetection |