EasyControlEdge: A Foundation-Model Fine-Tuning for Edge Detection

TL;DR

EasyControlEdge adapts image-generation foundation models for edge detection, achieving crisp, data-efficient results across various real-world applications by leveraging pretrained priors and iterative refinement.

Contribution

It introduces an edge-specialized adaptation of foundation models with an edge-oriented loss and guidance mechanism for improved edge detection performance.

Findings

Outperforms state-of-the-art methods on multiple datasets.

Achieves high crispness with limited training data.

Enables control over edge density during inference.

Abstract

We propose EasyControlEdge, adapting an image-generation foundation model to edge detection. In real-world edge detection (e.g., floor-plan walls, satellite roads/buildings, and medical organ boundaries), crispness and data efficiency are crucial, yet producing crisp raw edge maps with limited training samples remains challenging. Although image-generation foundation models perform well on many downstream tasks, their pretrained priors for data-efficient transfer and iterative refinement for high-frequency detail preservation remain underexploited for edge detection. To enable crisp and data-efficient edge detection using these capabilities, we introduce an edge-specialized adaptation of image-generation foundation models. To better specialize the foundation model for edge detection, we incorporate an edge-oriented objective with an efficient pixel-space loss. At inference, we introduce guidance based on unconditional dynamics, enabling a single model to control the edge density through a guidance scale. Experiments on BSDS500, NYUDv2, BIPED, and CubiCasa compare against state-of-the-art methods and show consistent gains, particularly under no-post-processing crispness evaluation and with limited training data.