# HED Edge Detection Canonical page: https://fuser.studio/models/hed-edge-detection HED Edge Detection in Fuser turns an image into a soft line map. Feed it to a ControlNet node to restyle an image while keeping its contours. ## Specs | Spec | Value | | --- | --- | | Inputs | Image | | Output | image | | Cost | 5 credits per run | Transform photographic boundaries into organic, sketch-like soft edge maps for flexible ControlNet guidance Creator: Saining Xie & Zhuowen Tu (HED) / ControlNet Community Also known as: HED, Holistically-Nested Edge Detection, Soft Edge [Extract Soft Edges](https://app.fuser.studio) ## How HED Edge Detection works Convert raw source photography into clean, expressive soft-edge conditioning maps in three direct actions. ### Upload Source Image Upload a high-contrast photograph or render with clearly defined subject outlines and strong foreground separation. ### Configure Edge Mode Toggle Safe Mode to eliminate background noise or Scribble Mode to generate bold felt-pen doodle contours. ### Extract Boundary Map Generate a luminous soft-edge boundary map ready to feed into Soft Edge or Scribble ControlNet pipelines. ## What HED Edge Detection is good at Explore the deep learning boundary capabilities that make HED the premier soft-edge preprocessor for generative conditioning. ### Organic Contour Extraction Captures continuous, human-like sketch lines with varied opacity and pen pressure across anatomical curves and natural contours. ### Safe Mode Noise Suppression Purges residual background grayscale noise to prevent downstream ControlNet models from hallucinating unwanted source textures or color casts. ### Scribble Mode Ink Simulation Compresses and thickens boundary lines into bold, tactile ink strokes designed for seamless integration with Scribble ControlNet workflows. ### Holistic Structural Preservation Preserves the macro volume, posture, and spatial depth of subjects while leaving micro-textures open for downstream stylistic reinterpretation. ## Made with HED Edge Detection See the structural fidelity and organic line flow achievable when guiding downstream renders with HED edge maps. ### Editorial fashion composition with organic drapery preservation ### Industrial concept rendering guided by soft structural contours ### Cinematic character portrait with expressive rim lighting ### Minimalist product hero shot with distinct silhouette definition ## What people build with HED Edge Detection Discover how art directors, concept artists, and visual designers use HED to steer complex image generation without rigid geometric lock-in. ### Fashion Lookbook Direction Retain complex fabric folds, tailored silhouettes, and dynamic model poses while giving diffusion models freedom over garment textile patterns and lighting. ### Cinematic Storyboarding Translate live-action frame captures into expressive sketch guides to direct atmospheric cinematic shots with precise character framing. ### Advertising Campaign Layouts Extract brand asset contours to generate bold advertising spreads that preserve product geometry while adapting surrounding art styles. ### Industrial Design Ideation Guide industrial concept renders from early clay models or foam prototypes without locking in clinical CAD wireframe stiffness. ### Album Art & Poster Design Generate thick, stylized contour guides with Scribble Mode to produce striking album sleeves and graphic print media. ## Frequently Asked Questions ### What is HED edge detection? HED stands for Holistically-Nested Edge Detection, a deep learning architecture developed by Saining Xie and Zhuowen Tu that uses fully convolutional networks for image boundary extraction. In generative AI workflows, HED is the standard preprocessor for Soft Edge ControlNet pipelines, delivering organic, sketch-like contour maps rather than rigid geometric vectors. ### How does HED differ from Canny edge detection? HED produces soft, continuous grayscale lines with varied pen-pressure gradients that mimic hand-drawn artist sketches, whereas Canny produces hard, binary, single-pixel lines. Reach for HED when preserving organic silhouettes, drapery, and human anatomy while giving downstream diffusion models freedom over fine texture; reach for Canny when you need rigid architectural wireframes or precise CAD-like borders. ### When should I enable safe mode in HED? Safe mode filters out faint background grayscale textures and subtle image noise from the source photo before generating the final boundary map. Enable safe mode when feeding downstream Soft Edge ControlNet models to prevent ghosting artifacts, color bleed, or texture hallucinations from leaking into your final render. ### What does scribble mode do in HED? Scribble mode binarizes and thickens the extracted contours to simulate bold felt-pen doodles and hand-drawn ink sketches. Use scribble mode when pairing with Scribble ControlNet checkpoints to achieve punchy graphic outlines, but leave it disabled for standard Soft Edge models to preserve delicate tonal gradients. ### What types of imagery should I avoid processing with HED? HED excels at broad silhouettes, body poses, and organic flow, but it struggles with pixel-perfect typography, intricate CAD blueprints, and high-frequency micro-textures like fine hair strands or tight fabric weaves. For technical vector alignment or micro-geometry, use Canny or Lineart preprocessors instead. ## Try HED Edge Detection on Fuser Transform photographic boundaries into organic, sketch-like soft edge maps for flexible ControlNet guidance [Extract Soft Edges](https://app.fuser.studio) ## Related models - [Gemini Image](https://fuser.studio/models/gemini-image.md) - [GPT Image](https://fuser.studio/models/gpt-image.md) - [Seedream](https://fuser.studio/models/seedream.md) - [FLUX.3 Image](https://fuser.studio/models/flux-3-image.md) - [FLUX.2](https://fuser.studio/models/flux-2.md) - [ESRGAN Upscaler](https://fuser.studio/models/esrgan-upscaler.md) Node reference: [HED Edge Detection docs](https://docs.fuser.studio/docs/nodes/image/hed-edge-detection.md)