Deep Image Segmentation via Discriminant Feature Learning

TL;DR

This paper introduces Deep Discriminant Analysis (DDA), a novel loss function that enhances image segmentation accuracy and boundary sharpness by embedding discriminant principles into training.

Contribution

The work proposes a differentiable, architecture-agnostic discriminant loss function that improves segmentation quality without increasing inference complexity.

Findings

DDA improves segmentation accuracy on DIS5K benchmark.

DDA enhances boundary sharpness and model confidence.

DDA is effective across various architectures.

Abstract

Accurate image segmentation remains challenging, particularly in generating sharp, confident boundaries. While modern architectures have advanced the field, many of them still rely on standard loss functions like Cross-Entropy and Dice, which often neglect the discriminative structure of learned features, leading to inaccurate boundaries. This work introduces Deep Discriminant Analysis (DDA), a differentiable, architecture-agnostic loss function that embeds classical discriminant principles for network training. DDA explicitly maximizes between-class variance while minimizing within-class one, promoting compact and separable feature distributions without increasing inference cost. Evaluations on the DIS5K benchmark demonstrate that DDA consistently improves segmentation accuracy, boundary sharpness, and model confidence across various architectures. Our results show that integrating discriminant analysis offers a simple, effective path for building more robust segmentation models.