Technical Papers

Perception

Tuesday, 23 July 9:00 AM - 10:30 AM
Session Chair: Diego Gutierrez, Universidad de Zaragoza

Exposing Photo Manipulation With Inconsistent Shadows

A forensic technique for determining if cast and attached shadows in an image are physically consistent. When inconsistencies are detected, the technique provides objective evidence of photo tampering.

Eric Kee
Dartmouth College

James O’Brien
University of California, Berkeley

Hany Farid
Dartmouth College

Gloss Perception in Painterly and Cartoon Rendering

This first study of material perception in stylized images (specifically painting and cartoon) uses non-photorealistic rendering algorithms to evaluate how such stylization alters the perception of gloss. The study estimates the function that maps realistic gloss parameters to their perception in a stylized rendering.

Adrien Bousseau
INRIA

James P. O'Shea
University of California, Berkeley

Frédo Durand
Massachusetts Institute of Technology CSAIL

Ravi Ramamoorthi
University of California, Berkeley

Maneesh Agrawala
University of California, Berkeley

Perception of Perspective Distortions in Image-Based Rendering

Street-level image-based rendering maps photographs of façades onto simple geometry, but viewing from an incorrect viewpoint can lead to misperceived geometry. This paper describes rigorous experiments that test competing hypotheses from vision science, develops a predictive model of perceived distortion, and provides guidelines for applications.

Peter Vangorp
REVES/INRIA, Sophia-Antipolis, University of Giessen, Max-Planck-Institut für Informatik

Christian Richardt
REVES/INRIA Sophia-Antipolis

Emily Cooper
University of California, Berkeley

Gaurav Chaurasia
REVES/INRIA Sophia Antipolis

Martin Banks
University of California, Berkeley

George Drettakis
REVES/INRIA Sophia-Antipolis

Understanding the Role of Phase Function in Translucent Appearance

This generalization of scattering phase-function models demonstrates an expanded translucent appearance space and discovers perceptually meaningful translucency controls by analyzing thousands of images with computation and psychophysics.

Ioannis Gkioulekas
Harvard School of Engineering and Applied Sciences

Bei Xiao
Massachusetts Institute of Technology

Shuang Zhao
Cornell University

Edward H. Adelson
Massachusetts Institute of Technology

Todd Zickler
Harvard School of Engineering and Applied Sciences

Kavita Bala
Cornell University