AI-Driven Deepfake Detection: A Sequential Learning Approach with LSTM

Radha Seelaboyina, Karnati Mysanthosh, K. Sri Vishnu Kshiraj, Sourav Kumar

2025

Abstract

The rapid advancement of deep learning has made video manipulation techniques, such as deepfakes, widely accessible. This research investigates the use of Long Short- Term Memory (LSTM) networks for detecting deepfake videos because of their capacity to learn temporal dependencies from different time intervals between neighboring frames. Differently from CNN-based approaches that focus on analyzing individual image frames in isolation, LSTM networks consider sequences of images, which helps normal video frames to reveal unnatural transitions and inconsistencies in the motion of facial components that are particular for deepfake videos. The proposed model induced both spatial features, which are artifacts appearing in a single frame, and temporal artifacts, which are inconsistencies among frames, through the integration of LSTM with CNNs, thus improving the reliability and precision of deepfake detection.

Paper Citation

in Harvard Style

Seelaboyina R., Mysanthosh K., Kshiraj K. and Kumar S. (2025). AI-Driven Deepfake Detection: A Sequential Learning Approach with LSTM. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 62-69. DOI: 10.5220/0013922600004919

in Bibtex Style

@conference{icrdicct`2525,
author={Radha Seelaboyina and Karnati Mysanthosh and K. Kshiraj and Sourav Kumar},
title={AI-Driven Deepfake Detection: A Sequential Learning Approach with LSTM},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={62-69},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013922600004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - AI-Driven Deepfake Detection: A Sequential Learning Approach with LSTM
SN - 978-989-758-777-1
AU - Seelaboyina R.
AU - Mysanthosh K.
AU - Kshiraj K.
AU - Kumar S.
PY - 2025
SP - 62
EP - 69
DO - 10.5220/0013922600004919
PB - SciTePress