Contextual Flow

Contextual Flow

Northwestern University

IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2009 [pdf]

Motivation

Matching based on local brightness is quite limited, because small changes on local appearance invalidate the constancy in brightness. The root of this limitation is its treatment regardless of the information from the spatial contexts. This paper presents a new approach that incorporates contexts to constrain motion estimation for target tracking. In this approach, one individual spatial context of a given pixel is represented by the posterior density of the associated feature class in its contextual domain. Each individual context gives a linear contextual flow constraint to the motion, so that the motion can be estimated in an over-determined contextual system. Based on this contextual flow model, this paper presents a new and powerful target tracking method that integrates the processes of salient contextual point selection, robust contextual matching, and dynamic context selection.

Contributions

There are mainly three contributions in our work

It goes from matching local brightness to matching spatial contexts for motion analysis.

As a counterpart of the optical flow constraint, contextual flow constraints are derived and introduced in this paper. Corresponding to a set of feature contexts, these contextual flow constraints are linear constraints so that the motion can be estimated in an over-determined linear system.

Based on contextual matching, this paper presents a new and powerful target tracking method that integrates processes of salient context point selection, robust contextual matching, and dynamic context selection.

Experimental Results

Click images to play the video. If video does not play, please install DivX video codec.


Giraffe.	Watch.

Return Home