Scalable Simple Linear Iterative Clustering (SSLIC) Using a Generic and Parallel Approach

TL;DR

This paper introduces a scalable, parallel implementation of the SLIC superpixel algorithm that efficiently handles n-dimensional images, demonstrating significant runtime improvements and scalability in processing complex datasets.

Contribution

It presents a generalized, multi-threaded version of the SLIC superpixel algorithm for n-dimensional images, enhancing scalability and performance over traditional implementations.

Findings

Good scalability demonstrated with large thread counts

Significant runtime gains on 3D and 2D datasets

Effective parallelization exceeds physical core limitations

Abstract

Superpixel algorithms have proven to be a useful initial step for segmentation and subsequent processing of images, reducing computational complexity by replacing the use of expensive per-pixel primitives with a higher-level abstraction, superpixels. They have been successfully applied both in the context of traditional image analysis and deep learning based approaches. In this work, we present a generalized implementation of the simple linear iterative clustering (SLIC) superpixel algorithm that has been generalized for n-dimensional scalar and multi-channel images. Additionally, the standard iterative implementation is replaced by a parallel, multi-threaded one. We describe the implementation details and analyze its scalability using a strong scaling formulation. Quantitative evaluation is performed using a 3D image, the Visible Human cryosection dataset, and a 2D image from the same dataset. Results show good scalability with runtime gains even when using a large number of threads that exceeds the physical number of available cores (hyperthreading).