Columbia Photographic Images and Photorealistic Computer Graphics Dataset Tian-Tsong Ng, Shih-Fu Chang, Jessie Hsu, Martin Pepeljugoski^¤ fttng,sfchang,yfhsug@ee.columbia.edu, [email protected] Department of Electrical Engineering Columbia University ADVENT Technical Report #205-2004-5 Feb 2005 Abstract Passive-blind image authentication is a new area of research.  A suitable dataset for experimentation and comparison of new techniques is important for the progress of the new research area.  In response to the need for a new dataset, the Columbia Photographic Images and Photorealistic Computer Graphics Dataset is made open for the passive-blind image authentication research community. The dataset is composed of four component image sets, i.e., the Photorealistic Com- puter Graphics  Set, the Personal Photographic Image  Set, the Google Image  Set, and the Recaptured Computer Graphics  Set. This dataset, available from http://www.ee.columbia.edu/trustfoto, will be for those who work on the photographic images versus photorealistic com- puter graphics classification problem, which is a subproblem of the passive-blind image authentication research.  In this report, we de- scribe the design and the implementation of the dataset. The report will also serve as a user guide for the dataset. 1   Introduction Digital watermarking [1] has been an active area of research since a decade ago. Various fragile [2, 3, 4, 5] or semi-fragile watermarking algorithms [6, 7, 8, 9] has been proposed for the image content authentication and the detection of image tampering.  In addition, authentication signature [10, ^¤This work was done when Martin spent his summer in our research group 1