Video Co-segmentation

This web page contains the results in video form of the paper

Video co-Segmentation
Jose C. Rubio, Joan Serrat, Antonio M. López
Asian Conference on Computer Vision 2012.

a PDF version of the paper can be found here

Abstract

Segmentation of a single image is in general a highly underconstrained problem. A frequent approach to solve it is to somehow provide prior knowledge or constraints on how the objects of interest look (in terms of their shape, size, color, location or structure). Image co-segmentation trades the need for such knowledge for something much easier to obtain, namely, additional images showing the object from other viewpoints. Now the segmentation problem is posed as one of differentiating the similar object regions in all the images from the more varying background.

In this paper, for the first time, we extend this approach to video segmentation: given two or more video sequences showing the same object (or objects belonging to the same class) moving in a similar manner, we aim to outline its region in all the frames. In addition, the method we present has two remarkable properties. First, it works in an unsupervised manner, by learning to segment at testing time, rather than requirement a training phase. Second, it produces not only a segmentation but also, as a byproduct, associations among corresponding space-time regions along the video sequences. Thus, if an object is formed by several parts with different aspect, as normally occurs, we are able to match them among the several sequences. We compare favorably with two state-of-the-art methods on video segmentation and report results on benchmark videos.

Sequences

ChaChaCha class

The category cha-cha-cha belongs to the Chroma dataset.

Cha-cha-cha1Cha-cha-cha2Cha-cha-cha3Cha-cha-cha4

This text will be replaced

IceSkater class

This video was presented in the paper Efficient Hierarchical Graph-Based Video Segmentation, by Grundmann et al.

Original videoYoutube1Youtube2

This text will be replaced

KiteSurfer class

This video was presented in the paper Efficient Hierarchical Graph-Based Video Segmentation, by Grundmann et al.

Original videoYoutube1Youtube2

This text will be replaced