Silhouette extraction from video System

In this project we represent a system that extracts an object silhouette from an input video.

The main goal of the project, is implementing the extraction algorithm described  on the article:
Shape-based Object Recognition in Videos Using 3D Synthetic Object Models on IEEE conference Computer Vision and Pattern Recognition, 2009. CVPR 2009

 Which addresses the problem of recognizing moving objects in videos by utilizing synthetic 3D models.  This paper presents the theoretical background and the implementation of the extraction algorithm. We extract object silhouettes from video by an integration of feature tracking, and co-segmentation of successive frames. The silhouette is constructed from the fusion of segments which contain the object’s feature points.