Learning Crisp Boundaries Using Deep Refinement Network and Adaptive Weighting Loss

TL;DR

This paper introduces DRNet, a deep refinement network with an adaptive loss function, achieving state-of-the-art results in boundary detection by focusing on crisp, precisely localized object contours.

Contribution

The paper proposes a novel deep refinement network and an adaptive loss function that combines cross-entropy and dice loss for improved boundary detection accuracy.

Findings

Achieved state-of-the-art performance on multiple datasets.

Analyzed and unified two methods for evaluating crisp boundaries.

Demonstrated the effectiveness of the adaptive loss in boundary localization.

Abstract

Significant progress has been made in boundary detection with the help of convolutional neural networks. Recent boundary detection models not only focus on real object boundary detection but also "crisp" boundaries (precisely localized along the object's contour). There are two methods to evaluate crisp boundary performance. One uses more strict tolerance to measure the distance between the ground truth and the detected contour. The other focuses on evaluating the contour map without any postprocessing. In this study, we analyze both methods and conclude that both methods are two aspects of crisp contour evaluation. Accordingly, we propose a novel network named deep refinement network (DRNet) that stacks multiple refinement modules to achieve richer feature representation and a novel loss function, which combines cross-entropy and dice loss through effective adaptive fusion. Experimental results demonstrated that we achieve state-of-the-art performance for several available datasets.