Date of Award

6-6-2015

Document Type

Thesis (Undergraduate)

Department or Program

Department of Computer Science

First Advisor

Hany Farid

Abstract

From its inception in 1960, computer graphics (CG) technology has quickly progressed from simple 3-D models to complex, photorealistic recreations of the human face and body. Alongside this innovation, lawmakers and courts in the United States have struggled to define what is illegal, what is "obscene”, and what is protected under the First Amendment with regards to child pornography. What has emerged from this debate is that the laws surrounding child pornography hinge on whether the material in question is photographic or CG. To this end, we measure how reliable the human visual system is in distinguishing CG from photographic images. After establishing a baseline for observer performance in this task as a function of both resolution and contrast, we address the following two questions: (1) is it possible to improve observer performance by isolating select features of the face? and (2) will training observers improve their performance?

Comments

Originally posted in the Dartmouth College Computer Science Technical Report Series, number TR2015-786.

Download
COinS