Conference 5–9 August 2012
Exhibition 7–9 August 2012
Los Angeles Convention Center

Fun With Video

Technical Papers

Fun With Video

Tuesday, 7 August 3:45 PM - 5:35 PM | Los Angeles Convention Center, Room 515AB
Session Chair: Noah Snavely, Cornell University

Video Deblurring for Hand-Held Cameras Using Patch-Based Synthesis

A video deblurring method for effectively restoring sharp frames from blurry ones caused by camera shakes. The method detects sharp regions in the video and uses them to restore blurry regions of the same content in nearby frames.

Sunghyun Cho
Pohang University of Science and Technolog

Jue Wang
Adobe Systems Incorporated

Seungyong Lee
Pohang University of Science and Technolog

Eulerian Video Magnification for Revealing Subtle Changes in the World

A method to temporally filter videos to reveal invisible variations in the scene. Applications include human pulse visualization and motion amplification.

Hao-Yu Wu
Massachusetts Institute of Technology

Michael Rubinstein
Massachusetts Institute of Technology

Eugene Shih
Quanta Research Cambridge, Inc.

John Guttag
Massachusetts Institute of Technology

Frédo Durand
Massachusetts Institute of Technology

William T. Freeman
Massachusetts Institute of Technology

Selectively De-Animating Video

This technique selectively removes large-scale motions of objects so that other motions are easier to see. Users draw three types of strokes indicating regions that should be de-animated, still, or dynamic in the output. The technique allows applications such as motion visualization, creation of cinemagraphs, and video editing.

Jiamin Bai
University of California, Berkeley

Aseem Agarwala
Adobe Systems Incorporated

Maneesh Agrawala
University of California, Berkeley

Ravi Ramamoorthi
University of California, Berkeley

Tools for Placing Cuts and Transitions in Interview Video

Tools designed to help editors place cuts and create transitions in interview video. The interface links a text transcript to the corresponding locations in the footage and renders cut suitability in the timeline. The system also generates visible or hidden seamless transitions to connect adjacent video segments.

Floraine Berthouzoz
University of California, Berkeley

Wilmot Li
Adobe Systems Incorporated

Maneesh Agrawala
University of California, Berkeley

Videoscapes: Exploring Sparse, Unstructured Video Collections

Videoscape is a data structure that enables interactive exploration of video collections by visually navigating (spatially and/or temporally) between different clips. The system leads to previously unseen ways of interactive spatio-temporal exploration of casually captured videos.

James Tompkin
University College London

Kwang In Kim
Max-Planck-Institut für Informatik

Jan Kautz
University College London

Christian Theobalt
Max-Planck-Institut für Informatik