Image Segmentation Keras : Implementation of Segnet, FCN, UNet, PSPNet and other models in Keras

TL;DR

This paper introduces a comprehensive Keras-based library implementing popular semantic segmentation models like SegNet, FCN, UNet, and PSPNet, and evaluates their performance across multiple datasets.

Contribution

It provides a unified implementation of key segmentation models in Keras along with comparative evaluation results, aiding researchers in model selection and development.

Findings

Models are implemented in Keras for ease of use.

Performance comparison across datasets is provided.

The library facilitates rapid experimentation with segmentation models.

Abstract

Semantic segmentation plays a vital role in computer vision tasks, enabling precise pixel-level understanding of images. In this paper, we present a comprehensive library for semantic segmentation, which contains implementations of popular segmentation models like SegNet, FCN, UNet, and PSPNet. We also evaluate and compare these models on several datasets, offering researchers and practitioners a powerful toolset for tackling diverse segmentation challenges.