TurboPixels: Fast Superpixels Using Geometric Flows

TLDR

The authors propose a geometric‑flow algorithm that generates dense superpixel oversegmentations of images. The method produces boundary‑respecting segments while enforcing a compactness constraint, achieving near‑linear time complexity that allows processing megapixel images with high superpixel densities in minutes. Qualitative tests on complex images and quantitative evaluation on the Berkeley dataset demonstrate that the algorithm yields high‑quality superpixels with reduced undersegmentation compared to non‑compact methods and achieves a significant speedup over N‑cuts.

Abstract

We describe a geometric-flow-based algorithm for computing a dense oversegmentation of an image, often referred to as superpixels. It produces segments that, on one hand, respect local image boundaries, while, on the other hand, limiting undersegmentation through a compactness constraint. It is very fast, with complexity that is approximately linear in image size, and can be applied to megapixel sized images with high superpixel densities in a matter of minutes. We show qualitative demonstrations of high-quality results on several complex images. The Berkeley database is used to quantitatively compare its performance to a number of oversegmentation algorithms, showing that it yields less undersegmentation than algorithms that lack a compactness constraint while offering a significant speedup over N-cuts, which does enforce compactness.