AI RESEARCH
MEMO: Human-like Crisp Edge Detection Using Masked Edge Prediction
arXiv CS.CV
•
ArXi:2603.20782v1 Announce Type: new Learning-based edge detection models trained with cross-entropy loss often suffer from thick edge predictions, which deviate from the crisp, single-pixel annotations typically provided by humans. While previous approaches to achieving crisp edges have focused on designing specialized loss functions or modifying network architectures, we show that a carefully designed