Video source attribution is commonly addressed by extracting the traces left into the content by the Photo Response Non-Uniformity (PRNU), originated by manufacturing processes in the form of slight imperfections in light response of pixels, and by comparing them to a reference trace characterizing the device. Results on the VISION dataset come up with some insights into optimizing video source attribution in different use cases. We, therefore, systematically analyze the PRNU contribute provided by all frames belonging to either digitally stabilized or not stabilized videos. In this paper, we explore whether an optimum strategy exists in selecting frames based on their type and their positions within the groups of pictures. This separated analysis makes it hard to understand if achieved conclusions still stand for digitally stabilized videos and if those choices represent a general optimum strategy to perform video source attribution. However, the two problems were always treated separately, and the combined effect of compression and digital stabilization was never considered. In the last decade, several approaches were proposed to overcome both these issues, mainly by selecting those video frames which are considered more informative. However, its effectiveness is mainly limited by compression and the effect of recently introduced electronic image stabilization on several devices. Photo Response Non-Uniformity (PRNU) is reputed the most successful trace to identify the source of a digital video.
0 Comments
Leave a Reply. |