Image forensics through illumination cues and deep learning

Abstract

This thesis proposes a number of forensics methods for detecting and localizing various types of forgeries present in images The first method focuses on exposing splicing forgeries involving human faces in the front pose It utilizes the inconsistencies in the lighting environments LEs in different faces present in the image under investigation A novel LE estimation method is proposed based on a low dimensional lighting model created from a set of front pose face images captured under diffe...