This repository serves as the course project for Advance in Computer Graphics, Zhejiang Unviersity. In this project, we have implemented the video object segmentation system describe in Video SnapCut: Robust Video Object Cutout Using Localized Classifiers.

In this project, we have implemented all the following:

• A complete pipeline of Video-SnapCut (Section 2-3 of the paper)

• Alpha matting (Section 4 of the paper) (We are using this implementation: AlphaMatting)

• UI for interactive refinement (Section 5 of the paper)

• Test and comparison against Adobe After Effects' Rotobrush, see our report for more details (Section 6 of the paper)

• OpenCV: 3.4.x
• QT: 5.x

You can build the project using QMake under several platforms. The Qt project file video-snapcut.pro is in the/video-snapcut-ui/video-snapcut folder. You can also use Qt Creator to open the project.

1. Xue Bai, Jue Wang, David Simons, and Guillermo Sapiro. 2009. Video SnapCut: robust video object cutout using localized classifiers.
2. Carsten Rother, Vladimir Kolmogorov, and Andrew Blake. 2004. "GrabCut": interactive foreground extraction using iterated graph cuts.
3. Y. Y. Boykov and M. -. Jolly. 2001. "Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images.