A Survey on Semi-Supervised Semantic Segmentation

TL;DR

This survey reviews recent semi-supervised semantic segmentation methods, categorizes them, and evaluates various models on benchmark datasets to identify current challenges and future research directions.

Contribution

It provides an updated taxonomy of semi-supervised segmentation methods and experimental comparison across categories on standard datasets.

Findings

Different methods show varying performance on benchmarks.

Semi-supervised approaches can significantly reduce labeling costs.

Challenges include class imbalance and domain adaptation.

Abstract

Semantic segmentation is one of the most challenging tasks in computer vision. However, in many applications, a frequent obstacle is the lack of labeled images, due to the high cost of pixel-level labeling. In this scenario, it makes sense to approach the problem from a semi-supervised point of view, where both labeled and unlabeled images are exploited. In recent years this line of research has gained much interest and many approaches have been published in this direction. Therefore, the main objective of this study is to provide an overview of the current state of the art in semi-supervised semantic segmentation, offering an updated taxonomy of all existing methods to date. This is complemented by an experimentation with a variety of models representing all the categories of the taxonomy on the most widely used becnhmark datasets in the literature, and a final discussion on the results obtained, the challenges and the most promising lines of future research.