DemonGatanjieuandClaude Opus 5.5
8b51dffd2b
OpenPose misread anime characters; DWPose (torch detector: the onnx one found nobody with this onnxruntime-gpu) reads a single anime character down to face and fingers. A crowded, heavily stylised picture can still come out empty, so both control blocks keep their map as a PreviewImage the AI sees in run_status's previews, and their next step says what to do when it is empty. Live on the isolated 8199 ComfyUI: DWPose followed a hand-behind-head pose (64 s); the three-elf picture stays unreadable for both readers. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
145 lines
4.1 KiB
JSON
145 lines
4.1 KiB
JSON
{
|
|
"job": "follow",
|
|
"rank": 2,
|
|
"short": "its layout",
|
|
"title": {
|
|
"zh": "照着构图画",
|
|
"en": "Follow a layout"
|
|
},
|
|
"about": {
|
|
"zh": "你递给 AI 一张图，AMB 读出远近和形状（深度图），让新图保持这个构图。 画布上是 Anima 模型时，自动改用 Anima 专用的 LLLite 控制网（需要 ComfyUI-Anima-LLLite）。",
|
|
"en": "From a picture you hand the AI, Anomalous reads its depth and keeps that layout and those shapes. On an Anima model it uses an Anima ControlNet-LLLite instead (needs ComfyUI-Anima-LLLite)."
|
|
},
|
|
"source": {
|
|
"zh": "参考工作流的「底图控制网」组（作者默认用深度）",
|
|
"en": "The reference workflow's ControlNet group (depth, its default)"
|
|
},
|
|
"packs": {
|
|
"DepthAnythingV2Preprocessor": "comfyui_controlnet_aux"
|
|
},
|
|
"takes": "reference",
|
|
"control": "depth",
|
|
"nodes": {
|
|
"reference": {
|
|
"class_type": "LoadImage",
|
|
"inputs": {
|
|
"image": "$image"
|
|
}
|
|
},
|
|
"depth": {
|
|
"class_type": "DepthAnythingV2Preprocessor",
|
|
"inputs": {
|
|
"image": [
|
|
"reference",
|
|
0
|
|
],
|
|
"ckpt_name": "depth_anything_v2_vitl.pth",
|
|
"resolution": 1024
|
|
}
|
|
},
|
|
"controlnet": {
|
|
"class_type": "ControlNetLoader",
|
|
"inputs": {
|
|
"control_net_name": "$controlnet"
|
|
}
|
|
},
|
|
"steer": {
|
|
"class_type": "ControlNetApplyAdvanced",
|
|
"inputs": {
|
|
"positive": [
|
|
"$sampler.positive",
|
|
"*"
|
|
],
|
|
"negative": [
|
|
"$sampler.negative",
|
|
"*"
|
|
],
|
|
"control_net": [
|
|
"controlnet",
|
|
0
|
|
],
|
|
"image": [
|
|
"depth",
|
|
0
|
|
],
|
|
"vae": [
|
|
"$decode.vae",
|
|
"*"
|
|
],
|
|
"strength": 0.7,
|
|
"start_percent": 0.0,
|
|
"end_percent": 0.7
|
|
},
|
|
"wire_to": "$sampler"
|
|
},
|
|
"look": {
|
|
"class_type": "PreviewImage",
|
|
"inputs": {
|
|
"images": [
|
|
"depth",
|
|
0
|
|
]
|
|
}
|
|
}
|
|
},
|
|
"after": "run_workflow: the picture now keeps that layout (strength 0.7, for the first 70% of the steps; lower either to let the model change more). The prompt still says what is drawn. Its depth map is among the run's previews: if it is empty or wrong (a crowded or very stylised picture), the picture is too hard to read.",
|
|
"variants": {
|
|
"anima": {
|
|
"packs": {
|
|
"DepthAnythingV2Preprocessor": "comfyui_controlnet_aux",
|
|
"AnimaLLLiteApply_sdscripts": "ComfyUI-Anima-LLLite"
|
|
},
|
|
"control": null,
|
|
"lllite": "depth",
|
|
"nodes": {
|
|
"reference": {
|
|
"class_type": "LoadImage",
|
|
"inputs": {
|
|
"image": "$image"
|
|
}
|
|
},
|
|
"depth": {
|
|
"class_type": "DepthAnythingV2Preprocessor",
|
|
"inputs": {
|
|
"image": [
|
|
"reference",
|
|
0
|
|
],
|
|
"ckpt_name": "depth_anything_v2_vitl.pth",
|
|
"resolution": 1024
|
|
}
|
|
},
|
|
"steer": {
|
|
"class_type": "AnimaLLLiteApply_sdscripts",
|
|
"inputs": {
|
|
"model": [
|
|
"$sampler.model",
|
|
"*"
|
|
],
|
|
"lllite_name": "$lllite",
|
|
"image": [
|
|
"depth",
|
|
0
|
|
],
|
|
"strength": 1.0,
|
|
"start_percent": 0.0,
|
|
"end_percent": 1.0,
|
|
"preserve_wrapper": true
|
|
},
|
|
"wire_to": "$sampler"
|
|
},
|
|
"look": {
|
|
"class_type": "PreviewImage",
|
|
"inputs": {
|
|
"images": [
|
|
"depth",
|
|
0
|
|
]
|
|
}
|
|
}
|
|
},
|
|
"after": "On this Anima model an Anima ControlNet-LLLite steers it, more gently than an SDXL ControlNet: say in the prompt what the picture shows (how many people, what they do), or the prompt wins. Strength 1.0 for all the steps; up to 1.5 holds it closer. run_workflow. Its depth map is among the run's previews: if it is empty or wrong (a crowded or very stylised picture), the picture is too hard to read."
|
|
}
|
|
}
|
|
}
|