Real And Deepfake Image Similarity Detection Using Transformers

DOI

10.2991/978-94-6463-858-5_28How to use a DOI?

Keywords

DeepFake; Image Similarity; Vision Transformers; AdamW optimizer

Abstract

This research proposes an innovative method for distinguishing real images from deepfakes using Vision Transformers (ViT) combined with the AdamW optimizer and cosine similarity. The approach utilizesViT’s self-attention mechanism to capture global spatial relationships and extract rich feature representations from input images. Cosine similarity is then used to measure the proximity between features extracted from real and deepfake images, enabling highly accurate deepfake detection. The AdamW optimizer facilitates effective training of the ViT by optimizing model parameters efficiently. Extensive evaluations across diverse scenarios demonstrate the robustness and precision of the proposed method, highlighting its potential in applications such as media verification, digital forensics, and surveillance.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - T. Abhinav
AU  - S. V. Vasantha
AU  - N. Prem Vardhan Naidu
AU  - G. Amulya
AU  - R. Srikanth
PY  - 2025
DA  - 2025/11/04
TI  - Real And Deepfake Image Similarity Detection Using Transformers
BT  - Proceedings of International Conference on Computer Science and Communication Engineering (ICCSCE 2025)
PB  - Atlantis Press
SP  - 317
EP  - 325
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-858-5_28
DO  - 10.2991/978-94-6463-858-5_28
ID  - Abhinav2025
ER  -