Abstract:Superpixel generation is increasingly an important area for computer vision tasks. While superpixels with highly regular shapes are preferred to make the subsequent processing easier, the accuracy of the superpixel boundaries is also necessary. Previous methods usually depend on a distance function considering both spatial and color coherency regularization on the whole image, which however is hard to balance between shape regularity and boundary adherence, especially when the desired number of superpixels is small. In addition, non-adaptive parameters and insufficient contour information also affect the performance of segmentation. To mitigate these problems, we propose a robust divide-and-conquer superpixel segmentation method, of which the core idea is that we apply a new contour information extraction and a pixel clustering to separate the input image into flat and non-flat regions, where the former targets shape regularity and the latter emphasizes boundary adherence, followed by an efficient hierarchical merging to clean up tiny and dangling superpixels. Our algorithm requires no additional parameter tuning except the desired number of superpixels since our internal parameters are self-adaptive to the image contents. Experimental results demonstrate that for public benchmark datasets, our algorithm consistently generates more regular superpixels with stronger boundary adherence than state-of-the-art methods while maintaining a competitive efficiency. The source code is available at https://github.com/YunyangXu/HQSGRD.

Watershed Superpixel

Decompose Image into Meaningful Regions Based on Contour Detector and Watershed Algorithm

An experimental comparison of superpixels detection methods for contour detection

Image Segmentation Based on Watershed Transform

An Improved Parallel Watershed Algorithm for Image Segmentation

Superpixel Segmentation Based on Square-Wise Asymmetric Partition and Structural Approximation.

Image Segmentation Using an Improved Watershed Algorithm

Fast and robust superpixel generation method

FOPWA： an Optimized Parallel Algorithm of Watershed Transform for Image Segmentation

New marker-based watershed algorithm

High Quality Superpixel Generation Through Regional Decomposition

Image Segmentation Method By Combining Watersheds And Ant Colony Clustering

A New Process for the Segmentation of High Resolution Remote Sensing Imagery

Superpixel Segmentation Based on Multiple Seed Growth

Adaptive Superpixel Segmentation Aggregating Local Contour and Texture Features

New Unsupervised Image Segmentation Via Marker-Based Watershed

Overflow Marker-based Watershed Segmentation Algorithm Based on Edge Detection

A Graph Based Superpixel Generation Algorithm

An Efficient Hybrid Linear Clustering Superpixel Decomposition Framework for Traffic Scene Semantic Segmentation.

Fast Superpixel Segmentation Via Boundary Sampling and Interpolation

Superpixel Segmentation: A Benchmark