SLRNet: Semi-Supervised Semantic Segmentation Via Label Reuse for Human Decomposition Images

TL;DR

SLRNet introduces a semi-supervised segmentation approach that reuses labels based on image similarities, improving performance on human decomposition images where labeled data is scarce.

Contribution

The paper presents a novel label reuse strategy tailored for semi-supervised segmentation in domain-specific datasets with limited annotations.

Findings

Outperforms state-of-the-art semi-supervised methods on human decomposition images

Effectively exploits similarities to reuse labels and improve segmentation accuracy

Demonstrates the approach's simplicity and effectiveness in a challenging domain

Abstract

Semantic segmentation is a challenging computer vision task demanding a significant amount of pixel-level annotated data. Producing such data is a time-consuming and costly process, especially for domains with a scarcity of experts, such as medicine or forensic anthropology. While numerous semi-supervised approaches have been developed to make the most from the limited labeled data and ample amount of unlabeled data, domain-specific real-world datasets often have characteristics that both reduce the effectiveness of off-the-shelf state-of-the-art methods and also provide opportunities to create new methods that exploit these characteristics. We propose and evaluate a semi-supervised method that reuses available labels for unlabeled images of a dataset by exploiting existing similarities, while dynamically weighting the impact of these reused labels in the training process. We evaluate our method on a large dataset of human decomposition images and find that our method, while conceptually simple, outperforms state-of-the-art consistency and pseudo-labeling-based methods for the segmentation of this dataset. This paper includes graphic content of human decomposition.