Technical Papers

Image & Video Editing

Thursday, 16 December | 4:15 PM - 6:30 PM | Room E1-E4

Presented in English / 영어로 발표 됨

Session Chair

Ariel Shamir

TOG Article: 145

Diffusion Maps for Edge-Aware Image Editing

This paper demonstrates the advantages of using diffusion distances instead of regular distances between pixel values in edge-aware image editing, and it analyzes efficient methods for computing such distances.

Zeev Farbman
The Hebrew University

Raanan Fattal
The Hebrew University

Dani Lischinski
The Hebrew University

TOG Article: 146

Data-Driven Image Color Theme Enhancement

Photographers carefully conceive the color composition of their images to convey desired color themes. This paper presents a data-driven method for enhancing a desired color theme in an image.

Baoyuan Wang
Zhejiang University

Yizhou Yu
University of Illinois at Urbana-Champaign

Tien-Tsin Wong
Chinese University of Hong Kong

Chun Chen
Zhejiang University

Ying-Qing Xu
Microsoft Research Asia

TOG Article: 147

Stereoscopic 3D Copy & Paste

This end-to-end system for 3D copy and paste handles occlusion and perspective change, and it preserves the stereo volume of copied objects.

Wan-Yen Lo
University of California, San Diego and Universität Bern

Matthias Zwicker
University of California, San Diego and Universität Bern

Markus Gross
ETH Zürich and Disney Research, Zürich

Jeroen van Baar
Disney Research Zürich

Claude Knaus
Universität Bern

TOG Article: 148

MovieReshape: Tracking and Reshaping of Humans in Videos

A method for quick and easy manipulation of the body shape and proportions of a human actor in arbitrary video footage.

Ajrun Jain
Max-Planck-Institut für Informatik

Thorsten Thormaehlen
Max-Planck-Institut für Informatik

Hans-Peter Seidel
Max-Planck-Institut für Informatik

Christian Theobalt
Max-Planck-Institut für Informatik

TOG Article: 149

Metric-Aware Processing of Spherical Imagery

An adapted finite-element system that supports intrinsic gradient-domain processing of spherical imagery.

Michael Kazhdan
The Johns Hopkins University

Hugues Hoppe
Microsoft Research