Abstract
Image edge detection (ED) faces a fundamental mismatch between training and inference: models are trained using continuous-valued outputs but evaluated using binary predictions. This misalignment, caused by the non-differentiability of binarization, weakens the link between learning objectives and actual task performance. In this paper, we propose a theoretical method to design a Binarization-Aware Adjuster (BAA), which explicitly incorporates binarization behavior into gradient-based optimization. At the core of BAA is a novel loss adjustment mechanism based on a Distance Weight Function (DWF), which reweights pixel-wise contributions according to their correctness and proximity to the decision boundary. This emphasizes decision-critical regions while down-weighting less influential ones. We also introduce a self-adaptive procedure to estimate the optimal binarization threshold for BAA, further aligning training dynamics with inference behavior. Extensive experiments across various architectures and datasets demonstrate the effectiveness of our approach. Beyond ED, BAA offers a generalizable strategy for bridging the gap between continuous optimization and discrete evaluation in structured prediction tasks.
Abstract (translated)
图像边缘检测（ED）在训练和推理之间存在根本性的不匹配：模型是使用连续值输出进行训练的，但在评估时却采用二元预测。这种不对齐是由二值化过程的非可微性所导致的，这削弱了学习目标与实际任务性能之间的联系。在这篇论文中，我们提出了一种理论方法来设计一种二值化感知调整器（BAA），该方法明确地将二值化行为整合到基于梯度的优化过程中。BAA的核心是一个新颖的损失调整机制，它基于距离权重函数（DWF）重新加权像素级别的贡献，根据其正确性和接近决策边界的程度来进行。 这种方法强调了在关键决策区域中的重要性，并降低了对影响较小部分的权重。我们还引入了一个自适应过程来估计BAA的最佳二值化阈值，进一步将训练动态与推理行为相协调。通过多种架构和数据集上的广泛实验，证明了我们的方法的有效性。除了ED之外，BAA为在结构化预测任务中弥合连续优化与离散评估之间的差距提供了一种通用策略。
URL
https://arxiv.org/abs/2506.12460
