Nodes/Forbidden Vision/Forbidden Vision 🎯 Fixer
ComfyUI Node

Forbidden Vision 🎯 Fixer

A ComfyUI node in Forbidden Vision with 35 inputs and 4 outputs.

By luxdelux7·Created about a year ago·Updated 23 days ago· 97
Forbidden Vision 🎯 Fixer
  • model
  • vae
  • positive
  • negative
  • image
  • latent
  • clip
  • final_image
  • processed_face
  • side_by_side_comparison
  • final_mask
steps10
cfg_scale3.0
samplereuler_ancestral
schedulersgm_uniform
denoise_strength0.40
seed0
face_selection0
detection_confidence0.75
manual_rotationNone
processing_resolution1024
enable_pre_upscaletrue
upscaler_modelFast 4x (Lanczos)
crop_padding1.6
face_positive_prompt
replace_positive_promptfalse
face_negative_prompt
replace_negative_promptfalse
exclusions
blend_softness8
mask_expansion2
sampling_mask_blur_size21
sampling_mask_blur_strength1.0
enable_color_correctiontrue
enable_segmentationtrue
enable_differential_diffusiontrue
enable_lightness_rescuetrue
enable_final_refinementtrue
offload_models_to_cputrue
CategoryForbidden Vision

Inputs (35)

NameTypeDefaultDescription
modelMODEL
vaeVAE
positiveCONDITIONING
negativeCONDITIONING
stepsINT101–100
cfg_scaleFLOAT3.00–30
samplerCOMBOeuler_ancestral44 options: euler, euler_cfg_pp, euler_ancestral, euler_ancestral_cfg_pp, heun, heunpp2, +38
schedulerCOMBOsgm_uniform9 options: simple, sgm_uniform, karras, exponential, ddim_uniform, beta, +3
denoise_strengthFLOAT0.400–1
seedINT00–18446744073709550000
face_selectionINT00–200=All faces, 1=1st face, etc.
detection_confidenceFLOAT0.750.1–1Face detection confidence threshold.
manual_rotationCOMBONoneManually rotate face crop before processing
processing_resolutionINT1024512–2048The resolution for processing.
enable_pre_upscaleBOOLEANtrueEnable to upscale small faces with an AI model before processing.
upscaler_modelCOMBOFast 4x (Lanczos)The model used for pre-upscaling small faces.
crop_paddingFLOAT1.61–3Padding added to the face region before inpaint.
face_positive_promptSTRING
replace_positive_promptBOOLEANfalse
face_negative_promptSTRING
replace_negative_promptBOOLEANfalse
exclusionsSTRINGWords/tags to remove from the main prompt specifically for the face processing step.
blend_softnessINT80–200
mask_expansionINT20–100
sampling_mask_blur_sizeINT211–101
sampling_mask_blur_strengthFLOAT1.00.1–6
enable_color_correctionBOOLEANtrue
enable_segmentationBOOLEANtrueUse AI segmentation. If disabled, creates oval masks.
enable_differential_diffusionBOOLEANtrueBetter blending. At high noise, the mask allows structure changes; at low noise, it locks the background.
enable_lightness_rescueBOOLEANtrueIf the generated face is darker than original, brighten it.
enable_final_refinementBOOLEANtrueRuns a quick 0.05 denoise pass at the end. Cleans artifacts and improves skin texture with sensitive models and higher denoise. Highly recommended.
offload_models_to_cpuBOOLEANtrueMove face detection/segmentation/corrector models from VRAM to RAM after processing. Frees GPU memory for other nodes (e.g. checkpoint swaps in queued workflows). Disable if you're chaining many face passes and want to skip the small reload overhead.
imageoptIMAGEOptional image input. If latent is also provided, latent will be used.
latentoptLATENTOptional latent input. Will be decoded for processing.
clipoptCLIPOptional: Required only if using face prompts.

Outputs (4)

NameTypeDescription
final_imageIMAGE
processed_faceIMAGE
side_by_side_comparisonIMAGE
final_maskMASK