MatchED: Crisp Edge Detection Using End-to-End, Matching-based Supervision

2026-02-24

AI Summary

Existing edge detection methods struggle to produce crisp, single-pixel-wide edges and often rely on non-differentiable post-processing steps such as non-maximum suppression (NMS), which hinders end-to-end optimization. To address this limitation, this work proposes MethodLPP, a lightweight plug-and-play module that enables end-to-end learning of sharp edges without requiring post-processing. By introducing a one-to-one matching supervision mechanism during training—based on spatial distance and confidence scores—the method achieves precise edge localization. With only approximately 21K parameters, MethodLPP significantly improves performance across four benchmark datasets, yielding 2–4× higher average clarity (AC), 20–35% gains in ODS, and notable improvements in OIS and AP, matching or surpassing state-of-the-art results obtained with conventional post-processing techniques.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Generating crisp, i.e., one-pixel-wide, edge maps remains one of the fundamental challenges in edge detection, affecting both traditional and learning-based methods. To obtain crisp edges, most existing approaches rely on two hand-crafted post-processing algorithms, Non-Maximum Suppression (NMS) and skeleton-based thinning, which are non-differentiable and hinder end-to-end optimization. Moreover, all existing crisp edge detection methods still depend on such post-processing to achieve satisfactory results. To address this limitation, we propose \MethodLPP, a lightweight, only $\sim$21K additional parameters, and plug-and-play matching-based supervision module that can be appended to any edge detection model for joint end-to-end learning of crisp edges. At each training iteration, \MethodLPP performs one-to-one matching between predicted and ground-truth edges based on spatial distance and confidence, ensuring consistency between training and testing protocols. Extensive experiments on four popular datasets demonstrate that integrating \MethodLPP substantially improves the performance of existing edge detection models. In particular, \MethodLPP increases the Average Crispness (AC) metric by up to 2--4$\times$ compared to baseline models. Under the crispness-emphasized evaluation (CEval), \MethodLPP further boosts baseline performance by up to 20--35\% in ODS and achieves similar gains in OIS and AP, achieving SOTA performance that matches or surpasses standard post-processing for the first time. Code is available at https://cvpr26-matched.github.io.

Problem

• crisp edge detection
• edge thinning
• non-maximum suppression
• end-to-end learning
• edge map

Innovation

• crisp edge detection
• end-to-end learning
• matching-based supervision
• non-maximum suppression
• edge thinning