FuSS: Fusing Superpixels for Improved Segmentation Consistency

TL;DR

This paper introduces FuSS, a superpixel-based post-processing method, and OpenGMM, a Gaussian Mixture Model extension, both enhancing semantic segmentation consistency and achieving state-of-the-art results on remote sensing datasets.

Contribution

The paper presents FuSS, a novel superpixel method, and extends OpenPCS with OpenGMM for multimodal pixel distribution modeling, improving segmentation accuracy.

Findings

FuSS achieves state-of-the-art results on ISPRS datasets.

Both proposed methods improve segmentation quality.

Post-processing with FuSS enhances quantitative and qualitative metrics.

Abstract

In this work, we propose two different approaches to improve the semantic consistency of Open Set Semantic Segmentation. First, we propose a method called OpenGMM that extends the OpenPCS framework using a Gaussian Mixture of Models to model the distribution of pixels for each class in a multimodal manner. The second approach is a post-processing which uses superpixels to enforce highly homogeneous regions to behave equally, rectifying erroneous classified pixels within these regions, we also proposed a novel superpixel method called FuSS. All tests were performed on ISPRS Vaihingen and Potsdam datasets, and both methods were capable to improve quantitative and qualitative results for both datasets. Besides that, the post-process with FuSS achieved state-of-the-art results for both datasets. The official implementation is available at: \url{https://github.com/iannunes/FuSS}.