A Survey of Semantic Segmentation

TL;DR

This survey provides a comprehensive overview of semantic segmentation techniques, including traditional methods and recent CNN-based approaches, along with evaluation metrics, datasets, and common challenges.

Contribution

It offers a structured taxonomy of segmentation algorithms and summarizes recent advances and evaluation practices in the field.

Findings

Traditional methods like Decision Forests and SVMs are discussed.

Recent CNN-based approaches are highlighted.

Common challenges in segmentation are examined.

Abstract

This survey gives an overview over different techniques used for pixel-level semantic segmentation. Metrics and datasets for the evaluation of segmentation algorithms and traditional approaches for segmentation such as unsupervised methods, Decision Forests and SVMs are described and pointers to the relevant papers are given. Recently published approaches with convolutional neural networks are mentioned and typical problematic situations for segmentation algorithms are examined. A taxonomy of segmentation algorithms is given.