FaceDetailer (pipe)

ImpactPack/Simple

FaceDetailerPipe

This node enhances details by automatically detecting specific objects in the input image using detection models (bbox, segm, sam) and regenerating the image by enlarging the detected area based on the guide size. Although this node is specialized to simplify the commonly used facial detail enhancement workflow, it can also be used for various automatic inpainting purposes depending on the detection model.

Example

JSON Example

{
  "class_type": "FaceDetailerPipe",
  "inputs": {
    "image": [
      "node_id",
      0
    ],
    "detailer_pipe": [
      "node_id",
      0
    ],
    "guide_size": 512,
    "guide_size_for": true,
    "max_size": 1024,
    "seed": 0,
    "steps": 20,
    "cfg": 8,
    "sampler_name": "euler",
    "scheduler": "simple",
    "denoise": 0.5,
    "feather": 5,
    "noise_mask": true,
    "force_inpaint": true,
    "bbox_threshold": 0.5,
    "bbox_dilation": 10,
    "bbox_crop_factor": 3,
    "sam_detection_hint": "center-1",
    "sam_dilation": 0,
    "sam_threshold": 0.93,
    "sam_bbox_expansion": 0,
    "sam_mask_hint_threshold": 0.7,
    "sam_mask_hint_use_negative": "False",
    "drop_size": 10,
    "refiner_ratio": 0.2,
    "cycle": 1
  }
}

This example shows required inputs only. Connection values like ["node_id", 0] should reference actual node IDs from your workflow.

Inputs

Name	Type	Status	Constraints	Default
image	IMAGE	required	-	-
detailer_pipe?	DETAILER_PIPE	required	-	-
guide_size	FLOAT	required	min: 64, max: 16384, step: 8	512
guide_size_for	BOOLEAN	required	-	true
max_size	FLOAT	required	min: 64, max: 16384, step: 8	1024
seed	INT	required	min: 0, max: 1.84e+19	0
steps	INT	required	min: 1, max: 10000	20
cfg	FLOAT	required	min: 0, max: 100	8
sampler_name	ENUM 44 options euler euler_cfg_pp euler_ancestral euler_ancestral_cfg_pp heun heunpp2 exp_heun_2_x0 exp_heun_2_x0_sde dpm_2 dpm_2_ancestral lms dpm_fast dpm_adaptive dpmpp_2s_ancestral dpmpp_2s_ancestral_cfg_pp dpmpp_sde dpmpp_sde_gpu dpmpp_2m dpmpp_2m_cfg_pp dpmpp_2m_sde dpmpp_2m_sde_gpu dpmpp_2m_sde_heun dpmpp_2m_sde_heun_gpu dpmpp_3m_sde dpmpp_3m_sde_gpu ddpm lcm ipndm ipndm_v deis res_multistep res_multistep_cfg_pp res_multistep_ancestral res_multistep_ancestral_cfg_pp gradient_estimation gradient_estimation_cfg_pp er_sde seeds_2 seeds_3 sa_solver sa_solver_pece ddim uni_pc uni_pc_bh2	required	-	-
scheduler	ENUM 17 options simple sgm_uniform karras exponential ddim_uniform beta normal linear_quadratic kl_optimal AYS SDXL AYS SD1 AYS SVD GITS[coeff=1.2] LTXV[default] OSS FLUX OSS Wan OSS Chroma	required	-	-
denoise	FLOAT	required	min: 0.0001, max: 1, step: 0.01	0.5
feather	INT	required	min: 0, max: 100, step: 1	5
noise_mask	BOOLEAN	required	-	true
force_inpaint	BOOLEAN	required	-	true
bbox_threshold	FLOAT	required	min: 0, max: 1, step: 0.01	0.5
bbox_dilation	INT	required	min: -512, max: 512, step: 1	10
bbox_crop_factor	FLOAT	required	min: 1, max: 10, step: 0.1	3
sam_detection_hint	ENUM 9 options center-1 horizontal-2 vertical-2 rect-4 diamond-4 mask-area mask-points mask-point-bbox none	required	-	-
sam_dilation	INT	required	min: -512, max: 512, step: 1	0
sam_threshold	FLOAT	required	min: 0, max: 1, step: 0.01	0.93
sam_bbox_expansion	INT	required	min: 0, max: 1000, step: 1	0
sam_mask_hint_threshold	FLOAT	required	min: 0, max: 1, step: 0.01	0.7
sam_mask_hint_use_negative	ENUM 3 options False Small Outter	required	-	-
drop_size	INT	required	min: 1, max: 16384, step: 1	10
refiner_ratio	FLOAT	required	min: 0, max: 1	0.2
cycle	INT	required	min: 1, max: 10, step: 1	1
inpaint_model	BOOLEAN	optional	-	false
noise_mask_feather	INT	optional	min: 0, max: 100, step: 1	20
scheduler_func_opt	SCHEDULER_FUNC	optional	-	-
tiled_encode	BOOLEAN	optional	-	false
tiled_decode	BOOLEAN	optional	-	false

Outputs

Index	Name	Type	Is List	Connection Reference
0	image	IMAGE	No	["{node_id}", 0]
1	cropped_refined	IMAGE	Yes	["{node_id}", 1]
2	cropped_enhanced_alpha	IMAGE	Yes	["{node_id}", 2]
3	mask	MASK	No	["{node_id}", 3]
4	detailer_pipe	DETAILER_PIPE	No	["{node_id}", 4]
5	cnet_images	IMAGE	Yes	["{node_id}", 5]

How to connect to these outputs

To connect another node's input to an output from this node, use the connection reference format:

["node_id", output_index]

Where node_id is the ID of this FaceDetailerPipe node in your workflow, and output_index is the index from the table above.

Example

If this node has ID "5" in your workflow:

• image (IMAGE): ["5", 0]
• cropped_refined (IMAGE): ["5", 1]
• cropped_enhanced_alpha (IMAGE): ["5", 2]

Package: comfyui-impact-pack