A survey of loss functions for semantic segmentation

TL;DR

This survey reviews various loss functions used in semantic segmentation, introduces a new log-cosh dice loss, and compares their performance across datasets to identify robust options for diverse scenarios.

Contribution

The paper summarizes existing loss functions, introduces a novel log-cosh dice loss, and evaluates their effectiveness, providing insights for better model convergence in semantic segmentation.

Findings

Certain loss functions perform consistently well across datasets.

The new log-cosh dice loss shows competitive performance.

Some loss functions can serve as reliable baselines in unknown data scenarios.

Abstract

Image Segmentation has been an active field of research as it has a wide range of applications, ranging from automated disease detection to self-driving cars. In the past five years, various papers came up with different objective loss functions used in different cases such as biased data, sparse segmentation, etc. In this paper, we have summarized some of the well-known loss functions widely used for Image Segmentation and listed out the cases where their usage can help in fast and better convergence of a model. Furthermore, we have also introduced a new log-cosh dice loss function and compared its performance on the NBFS skull-segmentation open-source data-set with widely used loss functions. We also showcased that certain loss functions perform well across all data-sets and can be taken as a good baseline choice in unknown data distribution scenarios. Our code is available at Github: https://github.com/shruti-jadon/Semantic-Segmentation-Loss-Functions.