Differential Diffusionとは？
通常の inpainting では、マスクの白い部分は変化し、黒い部分は変化しません。
では、グレーにすれば少しだけ変化するのか？というと、残念、想像するようなコントロールはそのままではできません。
そこで生まれたのが Differential Diffusion です。
マスクの濃度に応じて、場所ごとに denoise を変える ことが出来るため、場所ごとに変化量を変えたり、境界をぼかしたマスクを扱えるようになります。
inpainting をまだ読んでいなければ、先にご覧ください。
使い方
inpainting の workflow に Differential Diffusion ノードを追加し、マスクに濃淡をつけるだけです。
workflow
{
"id": "8b9f7796-0873-4025-be3c-0f997f67f866",
"revision": 0,
"last_node_id": 19,
"last_link_id": 23,
"nodes": [
{
"id": 12,
"type": "VAEEncode",
"pos": [
436.45541132111384,
672.3101875706969
],
"size": [
140,
46
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [
{
"name": "pixels",
"type": "IMAGE",
"link": 11
},
{
"name": "vae",
"type": "VAE",
"link": 13
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"links": [
14
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "VAEEncode"
},
"widgets_values": []
},
{
"id": 7,
"type": "CLIPTextEncode",
"pos": [
416.1970166015625,
392.37848510742185
],
"size": [
410.75801513671877,
158.82607910156253
],
"flags": {},
"order": 7,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 5
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
6
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"text, watermark, worst quality"
]
},
{
"id": 9,
"type": "SaveImage",
"pos": [
1451,
189
],
"size": [
503.2876035004722,
576.239673217884
],
"flags": {},
"order": 11,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 9
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33"
},
"widgets_values": [
"ComfyUI"
]
},
{
"id": 8,
"type": "VAEDecode",
"pos": [
1209,
188
],
"size": [
203.39999999999986,
46
],
"flags": {},
"order": 10,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 7
},
{
"name": "vae",
"type": "VAE",
"link": 10
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
9
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "VAEDecode"
},
"widgets_values": []
},
{
"id": 11,
"type": "LoadImage",
"pos": [
-46.552421344122806,
724.4939982393679
],
"size": [
348.6485235795455,
444.65055454545484
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
11
]
},
{
"name": "MASK",
"type": "MASK",
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "LoadImage",
"image": "clipspace/clipspace-painted-masked-1765071052914.png [input]"
},
"widgets_values": [
"8d2eb48340cf6f6f99e539e11517d6a2.png",
"image"
]
},
{
"id": 13,
"type": "SetLatentNoiseMask",
"pos": [
609.0212780548253,
672.3101875706969
],
"size": [
208.02038352272726,
46
],
"flags": {},
"order": 8,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 14
},
{
"name": "mask",
"type": "MASK",
"link": 23
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"links": [
16
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "SetLatentNoiseMask"
},
"widgets_values": []
},
{
"id": 10,
"type": "VAELoader",
"pos": [
167.95849892963756,
598.568575275356
],
"size": [
231.48760330578511,
58
],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "VAE",
"type": "VAE",
"links": [
10,
13
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "VAELoader"
},
"widgets_values": [
"vae-ft-mse-840000-ema-pruned.safetensors"
]
},
{
"id": 6,
"type": "CLIPTextEncode",
"pos": [
415,
186
],
"size": [
411.95503173828126,
151.0030493164063
],
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 3
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
4
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"RAW Photograph of a cross section of a kiwi for commercial use."
]
},
{
"id": 19,
"type": "LoadImageMask",
"pos": [
322.4453331961138,
798.0790526061061
],
"size": [
254.01007812500006,
368.578
],
"flags": {},
"order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MASK",
"type": "MASK",
"links": [
23
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "LoadImageMask"
},
"widgets_values": [
"d8cd78b75de91ed4e9a1da1eedfcf21d.png",
"red",
"image"
]
},
{
"id": 4,
"type": "CheckpointLoaderSimple",
"pos": [
39.900463636363625,
343.9095500000008
],
"size": [
315,
98
],
"flags": {},
"order": 3,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"slot_index": 0,
"links": [
21
]
},
{
"name": "CLIP",
"type": "CLIP",
"slot_index": 1,
"links": [
3,
5
]
},
{
"name": "VAE",
"type": "VAE",
"slot_index": 2,
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
"v1-5-pruned-emaonly-fp16.safetensors"
]
},
{
"id": 3,
"type": "KSampler",
"pos": [
863,
186
],
"size": [
315,
262
],
"flags": {},
"order": 9,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 22
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 4
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 6
},
{
"name": "latent_image",
"type": "LATENT",
"link": 16
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
7
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "KSampler"
},
"widgets_values": [
111111,
"fixed",
20,
8,
"euler",
"normal",
1
]
},
{
"id": 18,
"type": "DifferentialDiffusion",
"pos": [
526.0722281128822,
67.3403636363635
],
"size": [
210,
58
],
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 21
}
],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"links": [
22
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "DifferentialDiffusion"
},
"widgets_values": [
1
],
"color": "#232",
"bgcolor": "#353"
}
],
"links": [
[
3,
4,
1,
6,
0,
"CLIP"
],
[
4,
6,
0,
3,
1,
"CONDITIONING"
],
[
5,
4,
1,
7,
0,
"CLIP"
],
[
6,
7,
0,
3,
2,
"CONDITIONING"
],
[
7,
3,
0,
8,
0,
"LATENT"
],
[
9,
8,
0,
9,
0,
"IMAGE"
],
[
10,
10,
0,
8,
1,
"VAE"
],
[
11,
11,
0,
12,
0,
"IMAGE"
],
[
13,
10,
0,
12,
1,
"VAE"
],
[
14,
12,
0,
13,
0,
"LATENT"
],
[
16,
13,
0,
3,
3,
"LATENT"
],
[
21,
4,
0,
18,
0,
"MODEL"
],
[
22,
18,
0,
3,
0,
"MODEL"
],
[
23,
19,
0,
13,
1,
"MASK"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.620921323059155,
"offset": [
146.55242134412282,
24.607086363636487
]
},
"frontendVersion": "1.34.6",
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
"VHS_MetadataImage": true,
"VHS_KeepIntermediate": true
},
"version": 0.4
}
- 🟩
Differential Diffusionノードを追加 - ベースは
Set Latent Noise Maskノードを使った workflow です。- もちろん、inpainting モデルや ControlNet モデルを使った workflow でも使えます。
マスクの白い部分ほど大きく変化し、黒い部分ほど「元の絵」を残します。
マスクの使い方
場所ごとに変化量を変える
マスクの濃さは、滑らかなグラデーションにする必要はありません。
一枚のマスク画像の中で、場所ごとに濃さを変える ことで、1 回のサンプリングで部位ごとに違う変化量を指定できます。
{
"id": "8b9f7796-0873-4025-be3c-0f997f67f866",
"revision": 0,
"last_node_id": 19,
"last_link_id": 23,
"nodes": [
{
"id": 12,
"type": "VAEEncode",
"pos": [
436.45541132111384,
672.3101875706969
],
"size": [
140,
46
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [
{
"name": "pixels",
"type": "IMAGE",
"link": 11
},
{
"name": "vae",
"type": "VAE",
"link": 13
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"links": [
14
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "VAEEncode"
},
"widgets_values": []
},
{
"id": 7,
"type": "CLIPTextEncode",
"pos": [
416.1970166015625,
392.37848510742185
],
"size": [
410.75801513671877,
158.82607910156253
],
"flags": {},
"order": 7,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 5
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
6
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"text, watermark, worst quality"
]
},
{
"id": 9,
"type": "SaveImage",
"pos": [
1451,
189
],
"size": [
503.2876035004722,
576.239673217884
],
"flags": {},
"order": 11,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 9
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33"
},
"widgets_values": [
"ComfyUI"
]
},
{
"id": 8,
"type": "VAEDecode",
"pos": [
1209,
188
],
"size": [
203.39999999999986,
46
],
"flags": {},
"order": 10,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 7
},
{
"name": "vae",
"type": "VAE",
"link": 10
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
9
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "VAEDecode"
},
"widgets_values": []
},
{
"id": 11,
"type": "LoadImage",
"pos": [
-46.552421344122806,
724.4939982393679
],
"size": [
348.6485235795455,
444.65055454545484
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
11
]
},
{
"name": "MASK",
"type": "MASK",
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "LoadImage",
"image": "clipspace/clipspace-painted-masked-1765071052914.png [input]"
},
"widgets_values": [
"ff958820180efd9b316cb42ddd9c0276.png",
"image"
]
},
{
"id": 13,
"type": "SetLatentNoiseMask",
"pos": [
609.0212780548253,
672.3101875706969
],
"size": [
208.02038352272726,
46
],
"flags": {},
"order": 8,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 14
},
{
"name": "mask",
"type": "MASK",
"link": 23
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"links": [
16
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "SetLatentNoiseMask"
},
"widgets_values": []
},
{
"id": 10,
"type": "VAELoader",
"pos": [
167.95849892963756,
598.568575275356
],
"size": [
231.48760330578511,
58
],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "VAE",
"type": "VAE",
"links": [
10,
13
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "VAELoader"
},
"widgets_values": [
"vae-ft-mse-840000-ema-pruned.safetensors"
]
},
{
"id": 19,
"type": "LoadImageMask",
"pos": [
322.4453331961138,
798.0790526061061
],
"size": [
254.01007812500006,
368.578
],
"flags": {},
"order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MASK",
"type": "MASK",
"links": [
23
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "LoadImageMask"
},
"widgets_values": [
"2d0d14ad85109598f389e5ac0ad7b85f.png",
"red",
"image"
]
},
{
"id": 4,
"type": "CheckpointLoaderSimple",
"pos": [
39.900463636363625,
343.9095500000008
],
"size": [
315,
98
],
"flags": {},
"order": 3,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"slot_index": 0,
"links": [
21
]
},
{
"name": "CLIP",
"type": "CLIP",
"slot_index": 1,
"links": [
3,
5
]
},
{
"name": "VAE",
"type": "VAE",
"slot_index": 2,
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
"v1-5-pruned-emaonly-fp16.safetensors"
]
},
{
"id": 18,
"type": "DifferentialDiffusion",
"pos": [
526.0722281128822,
67.3403636363635
],
"size": [
210,
58
],
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 21
}
],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"links": [
22
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "DifferentialDiffusion"
},
"widgets_values": [
1
],
"color": "#232",
"bgcolor": "#353"
},
{
"id": 6,
"type": "CLIPTextEncode",
"pos": [
415,
186
],
"size": [
411.95503173828126,
151.0030493164063
],
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 3
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
4
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"high quality,high detailed,RAW Photograph of a cross section of a kiwi for commercial use.RAW Photograph of a cross section of a kiwi for commercial use."
]
},
{
"id": 3,
"type": "KSampler",
"pos": [
863,
186
],
"size": [
315,
262
],
"flags": {},
"order": 9,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 22
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 4
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 6
},
{
"name": "latent_image",
"type": "LATENT",
"link": 16
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
7
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "KSampler"
},
"widgets_values": [
3,
"fixed",
20,
8,
"euler",
"normal",
0.93
]
}
],
"links": [
[
3,
4,
1,
6,
0,
"CLIP"
],
[
4,
6,
0,
3,
1,
"CONDITIONING"
],
[
5,
4,
1,
7,
0,
"CLIP"
],
[
6,
7,
0,
3,
2,
"CONDITIONING"
],
[
7,
3,
0,
8,
0,
"LATENT"
],
[
9,
8,
0,
9,
0,
"IMAGE"
],
[
10,
10,
0,
8,
1,
"VAE"
],
[
11,
11,
0,
12,
0,
"IMAGE"
],
[
13,
10,
0,
12,
1,
"VAE"
],
[
14,
12,
0,
13,
0,
"LATENT"
],
[
16,
13,
0,
3,
3,
"LATENT"
],
[
21,
4,
0,
18,
0,
"MODEL"
],
[
22,
18,
0,
3,
0,
"MODEL"
],
[
23,
19,
0,
13,
1,
"MASK"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.683013455365071,
"offset": [
146.55242134412282,
32.659636363636494
]
},
"frontendVersion": "1.34.6",
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
"VHS_MetadataImage": true,
"VHS_KeepIntermediate": true
},
"version": 0.4
}
- 変えたい部分ごとに、マスクの濃さを描き分ける（例: 顔は薄めのグレー、背景は白など）
境界をなじませる
inpainting のよくある問題として、マスクの境目がくっきり出てしまうことがあります。
Differential Diffusion と、ぼかしたマスクを組み合わせて、境界を自然になじませましょう。
{
"id": "8b9f7796-0873-4025-be3c-0f997f67f866",
"revision": 0,
"last_node_id": 26,
"last_link_id": 37,
"nodes": [
{
"id": 9,
"type": "SaveImage",
"pos": [
1773.2999999999997,
213.20000000000002
],
"size": [
503.2876035004722,
576.239673217884
],
"flags": {},
"order": 12,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 9
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33"
},
"widgets_values": [
"ComfyUI"
]
},
{
"id": 8,
"type": "VAEDecode",
"pos": [
1531.2999999999997,
212.20000000000002
],
"size": [
203.39999999999986,
46
],
"flags": {},
"order": 11,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 7
},
{
"name": "vae",
"type": "VAE",
"link": 10
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
9
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "VAEDecode"
},
"widgets_values": []
},
{
"id": 7,
"type": "CLIPTextEncode",
"pos": [
416.1970166015625,
392.37848510742185
],
"size": [
410.75801513671877,
158.82607910156253
],
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 5
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
26
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"text, watermark, worst quality"
]
},
{
"id": 10,
"type": "VAELoader",
"pos": [
595.4674284324961,
604.8385752753561
],
"size": [
231.48760330578511,
58
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "VAE",
"type": "VAE",
"links": [
10,
27
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "VAELoader"
},
"widgets_values": [
"vae-ft-mse-840000-ema-pruned.safetensors"
]
},
{
"id": 23,
"type": "GrowMask",
"pos": [
383.9960038422341,
782.5591360982014
],
"size": [
210,
82
],
"flags": {},
"order": 3,
"mode": 0,
"inputs": [
{
"name": "mask",
"type": "MASK",
"link": 31
}
],
"outputs": [
{
"name": "MASK",
"type": "MASK",
"links": [
33
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "GrowMask"
},
"widgets_values": [
10,
true
],
"color": "#432",
"bgcolor": "#653"
},
{
"id": 26,
"type": "DifferentialDiffusion",
"pos": [
911.9888354376291,
91.96613864265017
],
"size": [
210,
58
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 36
}
],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"links": [
37
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "DifferentialDiffusion"
},
"widgets_values": [
1
],
"color": "#232",
"bgcolor": "#353"
},
{
"id": 6,
"type": "CLIPTextEncode",
"pos": [
415,
186
],
"size": [
411.95503173828126,
151.0030493164063
],
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 3
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
25
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"high quality, RAW photo of a glass of cola. A glass with water droplets on it.\n"
]
},
{
"id": 3,
"type": "KSampler",
"pos": [
1185.2999999999995,
210.20000000000002
],
"size": [
315,
262
],
"flags": {},
"order": 10,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 37
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 23
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 24
},
{
"name": "latent_image",
"type": "LATENT",
"link": 28
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
7
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "KSampler"
},
"widgets_values": [
1234,
"fixed",
20,
8,
"euler",
"normal",
1
]
},
{
"id": 25,
"type": "ImpactGaussianBlurMask",
"pos": [
616.9550317382811,
783.3663338181777
],
"size": [
210,
82
],
"flags": {},
"order": 7,
"mode": 0,
"inputs": [
{
"name": "mask",
"type": "MASK",
"link": 33
}
],
"outputs": [
{
"name": "MASK",
"type": "MASK",
"links": [
32,
34
]
}
],
"properties": {
"cnr_id": "comfyui-impact-pack",
"ver": "61bd8397a18e7e7668e6a24e95168967768c2bed",
"Node name for S&R": "ImpactGaussianBlurMask"
},
"widgets_values": [
30,
10
],
"color": "#432",
"bgcolor": "#653"
},
{
"id": 24,
"type": "MaskPreview",
"pos": [
859.5925016752453,
786.4859030229644
],
"size": [
296.28740796393686,
258
],
"flags": {},
"order": 8,
"mode": 0,
"inputs": [
{
"name": "mask",
"type": "MASK",
"link": 32
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "MaskPreview"
},
"widgets_values": []
},
{
"id": 22,
"type": "LoadImage",
"pos": [
12.38845236664146,
714.6602606311046
],
"size": [
348.6485235795455,
444.65055454545484
],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
35
]
},
{
"name": "MASK",
"type": "MASK",
"links": [
31
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "LoadImage",
"image": "clipspace/clipspace-painted-masked-1765087764033.png [input]"
},
"widgets_values": [
"clipspace/clipspace-painted-masked-1765087764033.png [input]",
"image"
]
},
{
"id": 4,
"type": "CheckpointLoaderSimple",
"pos": [
39.900463636363625,
349.9095500000008
],
"size": [
315,
98
],
"flags": {},
"order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"slot_index": 0,
"links": [
36
]
},
{
"name": "CLIP",
"type": "CLIP",
"slot_index": 1,
"links": [
3,
5
]
},
{
"name": "VAE",
"type": "VAE",
"slot_index": 2,
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
"sd-v1-5-inpainting.ckpt"
],
"color": "#323",
"bgcolor": "#535"
},
{
"id": 20,
"type": "InpaintModelConditioning",
"pos": [
885.5197492645148,
229.90120706804433
],
"size": [
258.49395533092,
138
],
"flags": {},
"order": 9,
"mode": 0,
"inputs": [
{
"name": "positive",
"type": "CONDITIONING",
"link": 25
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 26
},
{
"name": "vae",
"type": "VAE",
"link": 27
},
{
"name": "pixels",
"type": "IMAGE",
"link": 35
},
{
"name": "mask",
"type": "MASK",
"link": 34
}
],
"outputs": [
{
"name": "positive",
"type": "CONDITIONING",
"links": [
23
]
},
{
"name": "negative",
"type": "CONDITIONING",
"links": [
24
]
},
{
"name": "latent",
"type": "LATENT",
"links": [
28
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "InpaintModelConditioning"
},
"widgets_values": [
true
],
"color": "#323",
"bgcolor": "#535"
}
],
"links": [
[
3,
4,
1,
6,
0,
"CLIP"
],
[
5,
4,
1,
7,
0,
"CLIP"
],
[
7,
3,
0,
8,
0,
"LATENT"
],
[
9,
8,
0,
9,
0,
"IMAGE"
],
[
10,
10,
0,
8,
1,
"VAE"
],
[
23,
20,
0,
3,
1,
"CONDITIONING"
],
[
24,
20,
1,
3,
2,
"CONDITIONING"
],
[
25,
6,
0,
20,
0,
"CONDITIONING"
],
[
26,
7,
0,
20,
1,
"CONDITIONING"
],
[
27,
10,
0,
20,
2,
"VAE"
],
[
28,
20,
2,
3,
3,
"LATENT"
],
[
31,
22,
1,
23,
0,
"MASK"
],
[
32,
25,
0,
24,
0,
"MASK"
],
[
33,
23,
0,
25,
0,
"MASK"
],
[
34,
25,
0,
20,
4,
"MASK"
],
[
35,
22,
0,
20,
3,
"IMAGE"
],
[
36,
4,
0,
26,
0,
"MODEL"
],
[
37,
26,
0,
3,
0,
"MODEL"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.7513148009015783,
"offset": [
87.61154763335854,
8.033861357349835
]
},
"frontendVersion": "1.34.6",
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
"VHS_MetadataImage": true,
"VHS_KeepIntermediate": true
},
"version": 0.4
}
- 🟪 今回は inpainting モデルを使った workflow に組み込みます。
- 🟨
Gaussian Blur Maskノード（ComfyUI-Impact-Pack）でマスクの境界をぼかす- ぼかすと実質マスクが小さくなるため、前処理として少しマスクを大きくしておきます。
深度マップをマスクとして使う
深度マップは白黒のグラデーションで表されます。
つまり、そのまま Differential Diffusion のマスクとして使うことが出来ます。
{
"id": "8b9f7796-0873-4025-be3c-0f997f67f866",
"revision": 0,
"last_node_id": 30,
"last_link_id": 50,
"nodes": [
{
"id": 8,
"type": "VAEDecode",
"pos": [
1531.2999999999997,
212.20000000000002
],
"size": [
203.39999999999986,
46
],
"flags": {},
"order": 13,
"mode": 0,
"inputs": [
{
"name": "samples",
"type": "LATENT",
"link": 7
},
{
"name": "vae",
"type": "VAE",
"link": 10
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"slot_index": 0,
"links": [
9
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "VAEDecode"
},
"widgets_values": []
},
{
"id": 10,
"type": "VAELoader",
"pos": [
595.4674284324961,
604.8385752753561
],
"size": [
231.48760330578511,
58
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "VAE",
"type": "VAE",
"links": [
10,
27
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "VAELoader"
},
"widgets_values": [
"vae-ft-mse-840000-ema-pruned.safetensors"
]
},
{
"id": 26,
"type": "DifferentialDiffusion",
"pos": [
911.9888354376291,
91.96613864265017
],
"size": [
210,
58
],
"flags": {},
"order": 3,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 36
}
],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"links": [
37
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "DifferentialDiffusion"
},
"widgets_values": [
1
],
"color": "#232",
"bgcolor": "#353"
},
{
"id": 4,
"type": "CheckpointLoaderSimple",
"pos": [
39.900463636363625,
349.9095500000008
],
"size": [
315,
98
],
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "MODEL",
"type": "MODEL",
"slot_index": 0,
"links": [
36
]
},
{
"name": "CLIP",
"type": "CLIP",
"slot_index": 1,
"links": [
3,
5
]
},
{
"name": "VAE",
"type": "VAE",
"slot_index": 2,
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CheckpointLoaderSimple"
},
"widgets_values": [
"sd-v1-5-inpainting.ckpt"
],
"color": "#323",
"bgcolor": "#535"
},
{
"id": 9,
"type": "SaveImage",
"pos": [
1773.2999999999997,
213.20000000000002
],
"size": [
681.4876035004725,
577.339673217884
],
"flags": {},
"order": 14,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 9
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33"
},
"widgets_values": [
"ComfyUI"
]
},
{
"id": 22,
"type": "LoadImage",
"pos": [
-372.39154763335796,
567.1202606311042
],
"size": [
348.6485235795455,
444.65055454545484
],
"flags": {},
"order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
35,
38
]
},
{
"name": "MASK",
"type": "MASK",
"links": []
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "LoadImage",
"image": "clipspace/clipspace-painted-masked-1765087764033.png [input]"
},
"widgets_values": [
"pasted/image (46).png",
"image"
]
},
{
"id": 28,
"type": "ImageToMask",
"pos": [
242.37278607894345,
724.0115675875168
],
"size": [
210,
58
],
"flags": {},
"order": 7,
"mode": 0,
"inputs": [
{
"name": "image",
"type": "IMAGE",
"link": 39
}
],
"outputs": [
{
"name": "MASK",
"type": "MASK",
"links": [
46
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "ImageToMask"
},
"widgets_values": [
"red"
]
},
{
"id": 29,
"type": "InvertMask",
"pos": [
463.6676231181421,
724.9477989676396
],
"size": [
140,
26
],
"flags": {},
"order": 8,
"mode": 0,
"inputs": [
{
"name": "mask",
"type": "MASK",
"link": 46
}
],
"outputs": [
{
"name": "MASK",
"type": "MASK",
"links": [
48
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "InvertMask"
},
"widgets_values": []
},
{
"id": 20,
"type": "InpaintModelConditioning",
"pos": [
885.5197492645148,
229.90120706804433
],
"size": [
258.49395533092,
138
],
"flags": {},
"order": 11,
"mode": 0,
"inputs": [
{
"name": "positive",
"type": "CONDITIONING",
"link": 25
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 26
},
{
"name": "vae",
"type": "VAE",
"link": 27
},
{
"name": "pixels",
"type": "IMAGE",
"link": 35
},
{
"name": "mask",
"type": "MASK",
"link": 50
}
],
"outputs": [
{
"name": "positive",
"type": "CONDITIONING",
"links": [
23
]
},
{
"name": "negative",
"type": "CONDITIONING",
"links": [
24
]
},
{
"name": "latent",
"type": "LATENT",
"links": [
28
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "InpaintModelConditioning"
},
"widgets_values": [
true
],
"color": "#323",
"bgcolor": "#535"
},
{
"id": 24,
"type": "MaskPreview",
"pos": [
863.712501675244,
720.3897046758566
],
"size": [
296.28740796393686,
258
],
"flags": {},
"order": 10,
"mode": 0,
"inputs": [
{
"name": "mask",
"type": "MASK",
"link": 49
}
],
"outputs": [],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.76",
"Node name for S&R": "MaskPreview"
},
"widgets_values": []
},
{
"id": 7,
"type": "CLIPTextEncode",
"pos": [
416.1970166015625,
392.37848510742185
],
"size": [
410.75801513671877,
158.82607910156253
],
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 5
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
26
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"low quality, worst quality, blurry, out of focus, deformed face, distorted body, extra limbs, extra legs, extra arms, duplicate people, text, watermark, logo, frame, monochrome, black and white"
]
},
{
"id": 6,
"type": "CLIPTextEncode",
"pos": [
415,
186
],
"size": [
411.95503173828126,
151.0030493164063
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 3
}
],
"outputs": [
{
"name": "CONDITIONING",
"type": "CONDITIONING",
"slot_index": 0,
"links": [
25
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "CLIPTextEncode"
},
"widgets_values": [
"(best quality, high quality, detailed),beautifull woman walking to viewer, city transformed into a colorful floral scene, (vivid colors:1.1), blooming flowers growing from windows and walls,35mm lens, realistic lighting, ultra detailed, photorealistic"
]
},
{
"id": 30,
"type": "RemapMaskRange",
"pos": [
618.4611687276335,
725.6004563565681
],
"size": [
210,
82
],
"flags": {},
"order": 9,
"mode": 0,
"inputs": [
{
"name": "mask",
"type": "MASK",
"link": 48
}
],
"outputs": [
{
"name": "mask",
"type": "MASK",
"links": [
49,
50
]
}
],
"properties": {
"cnr_id": "comfyui-kjnodes",
"ver": "4dfb85dcc52e4315c33170d97bb987baa46d128b",
"Node name for S&R": "RemapMaskRange"
},
"widgets_values": [
-0.3,
1.6
]
},
{
"id": 27,
"type": "DepthAnythingV2Preprocessor",
"pos": [
2.3412432757320483,
721.9022295517397
],
"size": [
219.6646484375,
82
],
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "image",
"type": "IMAGE",
"link": 38
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
39
]
}
],
"properties": {
"cnr_id": "comfyui_controlnet_aux",
"ver": "12f35647f0d510e03b45a47fb420fe1245a575df",
"Node name for S&R": "DepthAnythingV2Preprocessor"
},
"widgets_values": [
"depth_anything_v2_vitl.pth",
512
],
"color": "#223",
"bgcolor": "#335"
},
{
"id": 3,
"type": "KSampler",
"pos": [
1185.2999999999995,
210.20000000000002
],
"size": [
315,
262
],
"flags": {},
"order": 12,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "MODEL",
"link": 37
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 23
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 24
},
{
"name": "latent_image",
"type": "LATENT",
"link": 28
}
],
"outputs": [
{
"name": "LATENT",
"type": "LATENT",
"slot_index": 0,
"links": [
7
]
}
],
"properties": {
"cnr_id": "comfy-core",
"ver": "0.3.33",
"Node name for S&R": "KSampler"
},
"widgets_values": [
987,
"fixed",
20,
8,
"euler",
"normal",
1
]
}
],
"links": [
[
3,
4,
1,
6,
0,
"CLIP"
],
[
5,
4,
1,
7,
0,
"CLIP"
],
[
7,
3,
0,
8,
0,
"LATENT"
],
[
9,
8,
0,
9,
0,
"IMAGE"
],
[
10,
10,
0,
8,
1,
"VAE"
],
[
23,
20,
0,
3,
1,
"CONDITIONING"
],
[
24,
20,
1,
3,
2,
"CONDITIONING"
],
[
25,
6,
0,
20,
0,
"CONDITIONING"
],
[
26,
7,
0,
20,
1,
"CONDITIONING"
],
[
27,
10,
0,
20,
2,
"VAE"
],
[
28,
20,
2,
3,
3,
"LATENT"
],
[
35,
22,
0,
20,
3,
"IMAGE"
],
[
36,
4,
0,
26,
0,
"MODEL"
],
[
37,
26,
0,
3,
0,
"MODEL"
],
[
38,
22,
0,
27,
0,
"IMAGE"
],
[
39,
27,
0,
28,
0,
"IMAGE"
],
[
46,
28,
0,
29,
0,
"MASK"
],
[
48,
29,
0,
30,
0,
"MASK"
],
[
49,
30,
0,
24,
0,
"MASK"
],
[
50,
30,
0,
20,
4,
"MASK"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.6830134553650705,
"offset": [
472.39154763335796,
8.033861357349835
]
},
"frontendVersion": "1.34.6",
"VHS_latentpreview": false,
"VHS_latentpreviewrate": 0,
"VHS_MetadataImage": true,
"VHS_KeepIntermediate": true
},
"version": 0.4
}
- 🟦 Depth Anything V2（comfyui_controlnet_aux）で深度マップを作成
- これは IMAGE なので
Convert Image to Maskノードでマスクに変換します。
- これは IMAGE なので
RemapMaskRange（ComfyUI-KJNodes）で濃淡を調整
正直、SD1.5 では性能が足りないのですが、深度マップをマスクとして使うこと自体は、お気に入りの方法です。
サンプル画像
