Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

Interactive Video Cutout Jue Wang1 Pravin Bhat1 R. Alex Colburn2 Maneesh Agrawala2 Michael F. Cohen2

Summary: Interactive Video Cutout
Jue Wang1 Pravin Bhat1 R. Alex Colburn2 Maneesh Agrawala2 Michael F. Cohen2
1University of Washington 2Microsoft Research
Input videos Cutout objects Foreground objects composited together on new background
Figure 1: Our interactive video cutout system makes it easy to extract the foreground objects in these videos of an elephant and a skatebaorder.
We then composite these objects onto a third background video to form a new video in which the skateboarder skates on the elephant.
We present an interactive system for efficiently extracting fore-
ground objects from a video. We extend previous min-cut based
image segmentation techniques to the domain of video with four
new contributions. We provide a novel painting-based user inter-
face that allows users to easily indicate the foreground object across
space and time. We introduce a hierarchical mean-shift preprocess
in order to minimize the number of nodes that min-cut must op-
erate on. Within the min-cut we also define new local cost func-
tions to augment the global costs defined in earlier work. Finally,
we extend 2D alpha matting methods designed for images to work
with 3D video volumes. We demonstrate that our matting approach
preserves smoothness across both space and time. Our interactive
video cutout system allows users to quickly extract foreground ob-


Source: Agrawala, Maneesh - Department of Electrical Engineering and Computer Sciences, University of California at Berkeley
Wang, Jue - Adobe
Washington at Seattle, University of - Department of Computer Science and Engineering, Graphics and Imaging Laboratory


Collections: Computer Technologies and Information Sciences