Deep Convolutional Neural Network for Graphics Forgery Detection in Video

• Published in: Wireless personal communications Vol. 112; no. 3; pp. 1763 - 1781
• Main Authors: Kaur, Harpreet, Jindal, Neeru
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2020; Springer Nature B.V
• Subjects: Abnormalities; Algorithms; Artificial neural networks; Communications Engineering; Computer Communication Networks; Computer simulation; Decoders; Engineering; Forgery; Frames (data processing); Machine learning; Networks; Neural networks; Signal,Image and Speech Processing; Accuracy; Forged frames; Deep convolutional neural network; Doctored video
• ISSN: 0929-6212; 1572-834X
• DOI: 10.1007/s11277-020-07126-3

Doctored video generation with easily accessible editing software has proved to be a major problem in maintaining its authenticity. Nowadays, deep neural networks have been recognized as an effective technique in eradicating such troubles and classify the abnormal variations in the videos directly by learning significant features. This white paper is focused on a highly efficient method for the exposure of inter-frame tampering in the videos by means of deep convolutional neural network (DCNN). The proposed algorithm will detect forgery without requiring additional pre-embedded information of the frame. The other significance of pre-existing learning-techniques is that our algorithm classifies the forged frames on the basis of the correlation between the frames and the observed abnormalities using DCNN. The decoders used for batch normalization of input improve the training swiftness. Simulation results obtained on the REWIND and GRIP video dataset with an average accuracy of 98% shows the superiority of the proposed algorithm as compared to the existing one. The efficiency of the proposed algorithm is also tested on YouTube compressed videos.