Skip to content

soumik12345/felzenszwalb_segmentation

Repository files navigation

Felzenszwalb Segmentation

Implementation of Efficient Graph-Based Image Segmentation by Pedro F. Felzenszwalb from Artificial Intelligence Lab, Massachusetts Institute of Technology and Daniel P. Huttenlocher from Computer Science Department, Cornell University. In this case, instead of treating image segmentation as a pixel-wise classification problem, it is defined as the task of partitioning a given image into a set of disjoint regions usually with a goal of simplifying the representation of the image into something that is more meaningful and easier to analyze.

Article: Efficient Graph-Based Image Segmentation

Usage

Install using pip install felzenszwalb-segmentation.

from felzenszwalb_segmentation import segment

segmented_image = segment(in_image, sigma, k, min_size)

Results