Showing 1–1 of 1 results for author: D'Avino, D
-
Autoencoder with recurrent neural networks for video forgery detection
Authors:
Dario D'Avino,
Davide Cozzolino,
Giovanni Poggi,
Luisa Verdoliva
Abstract:
Video forgery detection is becoming an important issue in recent years, because modern editing software provide powerful and easy-to-use tools to manipulate videos. In this paper we propose to perform detection by means of deep learning, with an architecture based on autoencoders and recurrent neural networks. A training phase on a few pristine frames allows the autoencoder to learn an intrinsic m…
▽ More
Video forgery detection is becoming an important issue in recent years, because modern editing software provide powerful and easy-to-use tools to manipulate videos. In this paper we propose to perform detection by means of deep learning, with an architecture based on autoencoders and recurrent neural networks. A training phase on a few pristine frames allows the autoencoder to learn an intrinsic model of the source. Then, forged material is singled out as anomalous, as it does not fit the learned model, and is encoded with a large reconstruction error. Recursive networks, implemented with the long short-term memory model, are used to exploit temporal dependencies. Preliminary results on forged videos show the potential of this approach.
△ Less
Submitted 29 August, 2017;
originally announced August 2017.
