Published Paper Details:

The growing computation power has made the deep learning algorithms so powerful thatcreating a indistinguishable human synthesized video popularly called as deep fakes havebecame very simple. Scenarios where these realistic face swapped deep fakes are used to createpolitical distress, fake terrorism events, revenge porn, blackmail peoples are easily envisioned.In this work, we describe a new deep learning-based method that can effectively distinguishAI-generated fake videos from real videos.Our method is capable of automatically detectingthe replacement and reenactment deep fakes. We are trying to use Artificial Intelligence(AI) tofight Artificial Intelligence(AI). Our system uses a Res-Next Convolution neural network toextract the frame-level features and these features and further used to train the Long Short TermMemory(LSTM) based Recurrent Neural Network(RNN) to classify whether the video issubject to any kind of manipulation or not, i.e whether the video is deep fake or real video. Toemulate the real time scenarios and make the model perform better on real time data, weevaluate our method on large amount of balanced and mixed data-set prepared by mixing thevarious available data-set like Face- Forensic++[1], Deepfake detection challenge[2], andCeleb-DF[3]. We also show how our system can achieve competitive result using very simpleand robust approach.