Graph Segmentation of Images

Posted on May 3, 2016


Use of graph representation of an image for segmentation. This is based on the following paper which is one of the most cited papers in Computer Vision. If you are starting to do research in computer vision related fields it is a good idea to understand few of those papers in as much detail as possible.

Felzenszwalb, Pedro F., and Daniel P. Huttenlocher. “Efficient graph-based image
segmentation.” International Journal of Computer Vision 59.2 (2004): 167-181.

The image pixels are modeled as graph nodes. Felzenzwalb proposed a graph theoretic greedy algorithm to determine evidence of a boundary and goes on to make sub-graphs denoting the segmentation of the image. It makes a very intelligent use of the data structure ‘Dis-joint Set’. The image below (right) shows the segmentation of the image (left). Code available at :

I have prepared detailed slides to understand this paper. This could be used to supplement the original publication for understanding the algorithm. Access slides HERE.

Screenshot from 2016-05-03 16:39:15.png

Original Manuscript : [Springer]
Project Page / Code : [More]
Supplementary Slides : [PDF]
(Slides prepared by Manohar Kuse based on Felzenszwalb et al’s paper)

Posted in: Research Blog