Technical Papers

Changing Your Perception

Thursday, 14 August 10:45 AM - 12:15 PM | Vancouver Convention Centre, East Building, Ballroom A Session Chair: Yaser Sheikh, Carnegie Mellon University

Modeling and Optimizing Eye Vergence Response to Stereoscopic Cuts

This paper describes a series of eye-tracking experiments to derive and evaluate a model describing adaptation of vergence to changes of disparity on a stereoscopic display. Estimated average observer parameters of the model were derived from the gathered data. This enables a range of strategies for minimizing the adaptation times.

Krzysztof Templin
CSAIL MIT, MPI Informatik

Piotr Didyk
CSAIL MIT

Karol Myszkowski
MPI Informatik

Mohamed M. Hefeeda
Qatar Computing Research Institute

Hans-Peter Seidel
MPI Informatik

Wojciech Matusik
MIT CSAIL

How Do People Edit Light Fields?

Investigating different interfaces to find the most convenient one for light field editing, based on the two main approaches used in existing literature. The approach includes two different experiments with these interfaces, where users are asked to perform several edits, on both synthetic and real light fields.

Adrian Jarabo
Universidad de Zaragoza

Belen Masia
Universidad de Zaragoza

Adrien Bousseau
REVES/INRIA Sophia-Antipolis

Fabio Pellacini
Sapienza - Università di Roma

Diego Gutierrez
Universidad de Zaragoza

Simulating and Compensating Changes in Appearance Between Day and Night Vision

This method retargets images from one luminance level to another while compensating for the differences in contrast and color perception. The method can compensate images on a dim display to appear as on a bright display, or simulate night vision for scene-refereed HDR images.

Robert Wanat
Bangor University

Rafal Mantiuk
Bangor University

Style Transfer for Headshot Portraits

This style-transfer method tailored for headshot portraits is local and multi-scale. The method is validated on a dataset containing 94 images from a web site.

YiChang Shih
CSAIL MIT

Sylvain Paris
Adobe Research

Connelly Barnes
University of Virginia

William Freeman
CSAIL MIT

Frédo Durand
CSAIL MIT

Transient Attributes for High-Level Understanding and Editing of Outdoor Scenes

This study of temporal variations in outdoor scene appearance annotates thousands of images from webcams and learns how to recognize high-level properties such as season or weather in new photographs. The paper shows how the method can be used for browsing photo collections and proposes a method for high-level image editing of outdoor scenes.

Pierre-Yves Laffont
Brown University

Zhile Ren
Brown University

Xiaofeng Tao
Brown University

Chao Qian
Brown University

James Hays
Brown University