Real or Fake?: Human judgments about photographs and computer-generated images of faces

Figure 1: A psychophysical experiment explored factors underlying human judgments of visual realism for face images. The figure shows subjective realism score of 10 photo-realistic computer-generated (CG) images and 10 photos of faces. CG images are above the scale and photos below it. Overall, photos were judged to be more realistic than CG images. While efforts to create realistic images have generated break-throughs in computer graphics modeling, there has been little re-search to date on factors causing people to perceive images as real versus computer-generated (CG). The motivation of the current ex-periment is to begin investigating such factors. We showed both computer-graphics experts and laypersons real photos and CG im-ages. Photo and CG images were of three types: original, modi-fied to show only intrinsic reflectance components, and modified to show only intrinsic shading components (grayscale). Participants judged whether each image was a real photo or a CG image. Re-sults showed that visual realism depends not only on image proper-ties, but also on cognitive characteristics of viewers. Shading was especially crucial for visual realism. Color was also important. Ex-perts outperformed laypersons, but their advantage was limited to grayscale images. This research at the interface between human cognition and computer vision is a starting point for investigating the factors underlying visual realism