Abstract: This article proposes a method for the motion segmentation using a moving camera. It classifies each image pixel in the image sequence as the background or the motion regions by applying a three-view constraint called the "parallax-based multiplanar constraint." The method is derived from the "Plane+Parallax" framework, it overcomes the problem of the previous geometry constraint and does not require the reference plane to be constant across multiple views. Meanwhile, the method modifies the surface degradation to the line degradation to detect the motion objects followed by a moving camera in the same direction. Experiments show the superiority of our method.

Key words: motion segmentation, geometry constraint, parallax, surface degradation