Deep Learning-Based Classification of Illumination Maps for Exposing Face Splicing Forgeries in Images

Abstract

This paper proposes a novel image forensics method which can detect splicing forgeries in human group portraits. The method converts an input image to an illumination map (IM), and the facial regions of the IM are compared in a pairwise manner using machine learning techniques to check the presence of splicing forgery. A siamese convolutional neural network (CNN) is first trained on an external training set to differentiate between face-IM pairs coming from similar and different illumination environments (IEs). Once trained, one of the twin CNNs of the siamese network is used as a feature extractor for each face present in a test image. The pairwise features are concatenated and classified using a support vector machine classifier for forgery detection. The advantage of the proposed method is its ability to learn features capable of differentiating faces coming from different IEs. The experimental results on multiple public datasets show the efficacy of the proposed method with respect to the state-of-the-art.