BONUS: ADETAILER

ADetailer (After Detailer) is an extension that automatically fixes faces and hands - the two things AI struggles with most.

WHAT IT DOES

After the main image generates, ADetailer:

Scans the image for faces (or hands, or other body parts)
Draws a box around each one it finds
Regenerates JUST that area with more detail
Blends the fixed version back into the image

It's like having an automatic touch-up artist.

WHY YOU MIGHT WANT IT

Even with good prompts and (symmetrical eyes)1.33, you'll still get:

One eye slightly larger than the other
Weird pupil shapes
Extra fingers or fused fingers
Blurry facial features in group shots

ADetailer catches these and fixes them automatically.

INSTALLING IT

In the A1111 WebUI:

Go to Extensions tab
Click "Available" sub-tab
Click "Load from" to refresh the list
Search for "adetailer"
Click Install
Go to "Installed" tab, click "Apply and restart UI"

Or install manually:

cd stable-diffusion-webui/extensions
git clone https://github.com/Bing-su/adetailer.git

THE DETECTION MODELS

ADetailer uses YOLO (You Only Look Once) models to find things. The main ones:

face_yolov8n.pt
  Finds faces. The "n" means nano - fast and lightweight.
  This is your default, works great.

face_yolov8s.pt
  The "s" means small - slightly more accurate, slightly slower.

hand_yolov8n.pt
  Finds hands. Useful but hands are harder to fix than faces.

person_yolov8n-seg.pt
  Finds whole people. Good for fixing body proportions.

Start with face_yolov8n. Add hand detection later if you're getting consistent hand problems.

OUR TESTED SETTINGS

These are the settings we actually use for our models:

ADetailer model: face_yolov8n.pt

ADetailer prompt:
  (perfect eyes:1.3), (detailed eyes:1.3), beautiful eyes,
  symmetrical eyes, detailed face, sharp focus

  This prompt tells ADetailer what to focus on when it regenerates
  the face area. We're specifically asking for better eyes.

Detection confidence: 0.3
  How sure it needs to be that it found a face.
  0.3 catches most faces without false positives.

Dilate/Erode: 4
  Expands the detection box slightly so it catches the whole face
  plus a little margin for blending.

Mask blur: 4
  Softens the edge of the inpainted area so it blends smoothly
  into the rest of the image.

Denoising strength: 0.4
  How much ADetailer can change the detected area.
  0.4 is moderate - fixes problems but doesn't dramatically
  alter the character's appearance.

Inpaint only masked: ON
  Only regenerates the detected face area, not the whole image.
  Leave this on.

Inpaint padding: 48
  Extra pixels around the detected area for context.
  Helps ADetailer understand what's around the face.

SEPARATE SETTINGS (ADVANCED)

ADetailer can use different generation settings than your main image. This is useful because face details benefit from different values:

Use separate steps: ON
ADetailer steps: 28
  More steps than some base generations, gives extra refinement
  to the face area.

Use separate CFG scale: ON
ADetailer CFG scale: 5.0
  Slightly lower CFG for more natural-looking facial features.

Use separate CLIP skip: ON
ADetailer CLIP skip: 2
  Matches our models (all use Clip Skip 2).

QUICK COPY-PASTE SETTINGS

If you just want to match our setup:

Model:              face_yolov8n.pt
Prompt:             (perfect eyes:1.3), (detailed eyes:1.3),
                    beautiful eyes, symmetrical eyes,
                    detailed face, sharp focus
Confidence:         0.3
Dilate/Erode:       4
Mask blur:          4
Denoising:          0.4
Inpaint only masked: ON
Inpaint padding:    48
Separate steps:     ON, 28 steps
Separate CFG:       ON, 5.0
Separate CLIP skip: ON, 2

WHEN TO USE IT

Good candidates:

Portraits where eyes need to be perfect
Images with hands visible (use hand_yolov8n.pt)
Group shots with multiple faces
Full body shots where faces are small

Skip it for:

Abstract art
Images without people
When you're just exploring (slows things down)

PERFORMANCE IMPACT

ADetailer adds processing time:

Detection pass: ~1-2 seconds
Per-face inpainting: ~10-20 seconds with our settings

If you're generating lots of test images, turn it off until you find a composition you like. Then enable it for the final render.

EXAMPLE WORKFLOW

1. Generate without ADetailer to find a good composition 2. Note the seed of an image you like 3. Enable ADetailer with our settings above 4. Regenerate with same seed 5. Compare - the face should be cleaner, especially eyes

If eyes or expression changed too much, lower the denoising strength to 0.3 or even 0.2.

STACKING DETECTORS

You can run multiple detectors in sequence:

1st ADetailer: face_yolov8n.pt (fixes face)
2nd ADetailer: hand_yolov8n.pt (then fixes hands)

Each one runs after the previous, so faces get fixed first, then hands get fixed on the face-corrected image.

Enable "2nd" in the ADetailer panel to add another detector.

COMMON ISSUES

"ADetailer keeps changing my character's expression"

Lower the denoising strength to 0.2-0.3

"It's not detecting faces"

Lower the confidence threshold to 0.2
Make sure faces are large enough in the image

"It's detecting things that aren't faces"

Raise the confidence threshold to 0.4-0.5

"My image looks patchy where it was inpainted"

Increase mask blur to 8-12 for smoother blending
Increase inpaint padding to 64

LINKS

GitHub: https://github.com/Bing-su/adetailer Documentation: Check the GitHub wiki for detailed settings