ComfyUI-CogVideoXWrapper/examples/cogvideox_Fun_pose_02.json

1121 lines
20 KiB
JSON

{
"last_node_id": 86,
"last_link_id": 195,
"nodes": [
{
"id": 60,
"type": "WidgetToString",
"pos": {
"0": 1084,
"1": 76
},
"size": {
"0": 315,
"1": 130
},
"flags": {},
"order": 5,
"mode": 0,
"inputs": [
{
"name": "any_input",
"type": "*",
"link": 128,
"shape": 7
}
],
"outputs": [
{
"name": "STRING",
"type": "STRING",
"links": [
129
],
"slot_index": 0,
"shape": 3
}
],
"properties": {
"Node name for S&R": "WidgetToString"
},
"widgets_values": [
0,
"prompt",
false,
""
]
},
{
"id": 59,
"type": "AddLabel",
"pos": {
"0": 1466,
"1": -5
},
"size": {
"0": 315,
"1": 274
},
"flags": {},
"order": 14,
"mode": 0,
"inputs": [
{
"name": "image",
"type": "IMAGE",
"link": 126
},
{
"name": "caption",
"type": "STRING",
"link": null,
"widget": {
"name": "caption"
}
},
{
"name": "text",
"type": "STRING",
"link": 129,
"widget": {
"name": "text"
}
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
150
],
"slot_index": 0,
"shape": 3
}
],
"properties": {
"Node name for S&R": "AddLabel"
},
"widgets_values": [
10,
2,
48,
12,
"white",
"black",
"FreeMono.ttf",
"Text",
"up",
""
]
},
{
"id": 44,
"type": "VHS_VideoCombine",
"pos": {
"0": 1842,
"1": -5
},
"size": [
1186.0863037109375,
1457.6950174967449
],
"flags": {},
"order": 15,
"mode": 0,
"inputs": [
{
"name": "images",
"type": "IMAGE",
"link": 150
},
{
"name": "audio",
"type": "AUDIO",
"link": null,
"shape": 7
},
{
"name": "meta_batch",
"type": "VHS_BatchManager",
"link": null,
"shape": 7
},
{
"name": "vae",
"type": "VAE",
"link": null,
"shape": 7
}
],
"outputs": [
{
"name": "Filenames",
"type": "VHS_FILENAMES",
"links": null,
"shape": 3
}
],
"properties": {
"Node name for S&R": "VHS_VideoCombine"
},
"widgets_values": {
"frame_rate": 8,
"loop_count": 0,
"filename_prefix": "CogVideoX_Fun_Pose",
"format": "video/h264-mp4",
"pix_fmt": "yuv420p",
"crf": 19,
"save_metadata": true,
"pingpong": false,
"save_output": true,
"videopreview": {
"hidden": false,
"paused": false,
"params": {
"filename": "CogVideoX_Fun_Pose_00001.mp4",
"subfolder": "",
"type": "temp",
"format": "video/h264-mp4",
"frame_rate": 8
},
"muted": false
}
}
},
{
"id": 84,
"type": "DownloadAndLoadCogVideoModel",
"pos": {
"0": 605,
"1": -12
},
"size": [
377.9334482359568,
218
],
"flags": {},
"order": 0,
"mode": 0,
"inputs": [
{
"name": "block_edit",
"type": "TRANSFORMERBLOCKS",
"link": null,
"shape": 7
},
{
"name": "lora",
"type": "COGLORA",
"link": null,
"shape": 7
},
{
"name": "compile_args",
"type": "COMPILEARGS",
"link": null,
"shape": 7
}
],
"outputs": [
{
"name": "model",
"type": "COGVIDEOMODEL",
"links": [
186
],
"slot_index": 0
},
{
"name": "vae",
"type": "VAE",
"links": [
188,
191
],
"slot_index": 1
}
],
"properties": {
"Node name for S&R": "DownloadAndLoadCogVideoModel"
},
"widgets_values": [
"alibaba-pai/CogVideoX-Fun-V1.1-5b-Pose",
"bf16",
"disabled",
false,
"sdpa",
"main_device"
]
},
{
"id": 85,
"type": "CogVideoDecode",
"pos": {
"0": 1461,
"1": 357
},
"size": {
"0": 315,
"1": 198
},
"flags": {},
"order": 12,
"mode": 0,
"inputs": [
{
"name": "vae",
"type": "VAE",
"link": 188
},
{
"name": "samples",
"type": "LATENT",
"link": 189
}
],
"outputs": [
{
"name": "images",
"type": "IMAGE",
"links": [
190
]
}
],
"properties": {
"Node name for S&R": "CogVideoDecode"
},
"widgets_values": [
true,
240,
360,
0.2,
0.2,
true
]
},
{
"id": 30,
"type": "CogVideoTextEncode",
"pos": {
"0": 513,
"1": 286
},
"size": {
"0": 471.90142822265625,
"1": 168.08047485351562
},
"flags": {},
"order": 3,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 54
}
],
"outputs": [
{
"name": "conditioning",
"type": "CONDITIONING",
"links": [
128,
183
],
"slot_index": 0,
"shape": 3
},
{
"name": "clip",
"type": "CLIP",
"links": [
194
],
"slot_index": 1
}
],
"properties": {
"Node name for S&R": "CogVideoTextEncode"
},
"widgets_values": [
"a brown bear is dancing in a forest, in front of a waterfall",
1,
false
]
},
{
"id": 31,
"type": "CogVideoTextEncode",
"pos": {
"0": 507,
"1": 517
},
"size": {
"0": 501.0985412597656,
"1": 144
},
"flags": {},
"order": 6,
"mode": 0,
"inputs": [
{
"name": "clip",
"type": "CLIP",
"link": 194
}
],
"outputs": [
{
"name": "conditioning",
"type": "CONDITIONING",
"links": [
184
],
"slot_index": 0,
"shape": 3
},
{
"name": "clip",
"type": "CLIP",
"links": null
}
],
"properties": {
"Node name for S&R": "CogVideoTextEncode"
},
"widgets_values": [
"The video is not of a high quality, it has a low resolution. Watermark present in each frame. Strange motion trajectory. Character is speaking",
1,
true
]
},
{
"id": 20,
"type": "CLIPLoader",
"pos": {
"0": 2,
"1": 412
},
"size": {
"0": 451.30548095703125,
"1": 82
},
"flags": {},
"order": 1,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "CLIP",
"type": "CLIP",
"links": [
54
],
"slot_index": 0,
"shape": 3
}
],
"properties": {
"Node name for S&R": "CLIPLoader"
},
"widgets_values": [
"t5\\google_t5-v1_1-xxl_encoderonly-fp8_e4m3fn.safetensors",
"sd3"
]
},
{
"id": 72,
"type": "INTConstant",
"pos": {
"0": -498,
"1": 276
},
"size": {
"0": 210,
"1": 58
},
"flags": {},
"order": 2,
"mode": 0,
"inputs": [],
"outputs": [
{
"name": "value",
"type": "INT",
"links": [
152,
195
],
"slot_index": 0,
"shape": 3
}
],
"title": "Frames",
"properties": {
"Node name for S&R": "INTConstant"
},
"widgets_values": [
49
],
"color": "#1b4669",
"bgcolor": "#29699c"
},
{
"id": 65,
"type": "VHS_LoadVideo",
"pos": {
"0": -506,
"1": 477
},
"size": [
390.1356201171875,
910.0188802083334
],
"flags": {},
"order": 4,
"mode": 0,
"inputs": [
{
"name": "meta_batch",
"type": "VHS_BatchManager",
"link": null,
"shape": 7
},
{
"name": "vae",
"type": "VAE",
"link": null,
"shape": 7
},
{
"name": "frame_load_cap",
"type": "INT",
"link": 152,
"widget": {
"name": "frame_load_cap"
}
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
173
],
"slot_index": 0,
"shape": 3
},
{
"name": "frame_count",
"type": "INT",
"links": null,
"shape": 3
},
{
"name": "audio",
"type": "AUDIO",
"links": null,
"shape": 3
},
{
"name": "video_info",
"type": "VHS_VIDEOINFO",
"links": [],
"slot_index": 3,
"shape": 3
}
],
"properties": {
"Node name for S&R": "VHS_LoadVideo"
},
"widgets_values": {
"video": "01.mp4",
"force_rate": 0,
"force_size": "Disabled",
"custom_width": 512,
"custom_height": 512,
"frame_load_cap": 17,
"skip_first_frames": 0,
"select_every_nth": 1,
"choose video to upload": "image",
"videopreview": {
"hidden": false,
"paused": false,
"params": {
"frame_load_cap": 17,
"skip_first_frames": 0,
"force_rate": 0,
"filename": "01.mp4",
"type": "input",
"format": "video/mp4",
"select_every_nth": 1
},
"muted": false
}
}
},
{
"id": 80,
"type": "DWPreprocessor",
"pos": {
"0": -66,
"1": 583
},
"size": {
"0": 364.7358703613281,
"1": 198
},
"flags": {},
"order": 7,
"mode": 0,
"inputs": [
{
"name": "image",
"type": "IMAGE",
"link": 173
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
174
],
"slot_index": 0,
"shape": 3
},
{
"name": "POSE_KEYPOINT",
"type": "POSE_KEYPOINT",
"links": null,
"shape": 3
}
],
"properties": {
"Node name for S&R": "DWPreprocessor"
},
"widgets_values": [
"enable",
"enable",
"enable",
512,
"yolox_l.torchscript.pt",
"dw-ll_ucoco_384_bs5.torchscript.pt"
]
},
{
"id": 37,
"type": "ImageResizeKJ",
"pos": {
"0": -4,
"1": 829
},
"size": {
"0": 315,
"1": 266
},
"flags": {},
"order": 8,
"mode": 0,
"inputs": [
{
"name": "image",
"type": "IMAGE",
"link": 174
},
{
"name": "get_image_size",
"type": "IMAGE",
"link": null,
"shape": 7
},
{
"name": "width_input",
"type": "INT",
"link": null,
"widget": {
"name": "width_input"
}
},
{
"name": "height_input",
"type": "INT",
"link": null,
"widget": {
"name": "height_input"
}
}
],
"outputs": [
{
"name": "IMAGE",
"type": "IMAGE",
"links": [
130
],
"slot_index": 0,
"shape": 3
},
{
"name": "width",
"type": "INT",
"links": null,
"shape": 3
},
{
"name": "height",
"type": "INT",
"links": null,
"shape": 3
}
],
"properties": {
"Node name for S&R": "ImageResizeKJ"
},
"widgets_values": [
512,
512,
"lanczos",
true,
16,
0,
0,
"disabled"
]
},
{
"id": 61,
"type": "GetImageSizeAndCount",
"pos": {
"0": 378,
"1": 828
},
"size": {
"0": 277.20001220703125,
"1": 86
},
"flags": {},
"order": 9,
"mode": 0,
"inputs": [
{
"name": "image",
"type": "IMAGE",
"link": 130
}
],
"outputs": [
{
"name": "image",
"type": "IMAGE",
"links": [
135,
192
],
"slot_index": 0,
"shape": 3
},
{
"name": "288 width",
"type": "INT",
"links": null,
"shape": 3
},
{
"name": "512 height",
"type": "INT",
"links": null,
"shape": 3
},
{
"name": "49 count",
"type": "INT",
"links": null,
"shape": 3
}
],
"properties": {
"Node name for S&R": "GetImageSizeAndCount"
},
"widgets_values": []
},
{
"id": 86,
"type": "CogVideoImageEncode",
"pos": {
"0": 717,
"1": 808
},
"size": {
"0": 315,
"1": 122
},
"flags": {},
"order": 10,
"mode": 0,
"inputs": [
{
"name": "vae",
"type": "VAE",
"link": 191
},
{
"name": "start_image",
"type": "IMAGE",
"link": 192
},
{
"name": "end_image",
"type": "IMAGE",
"link": null,
"shape": 7
}
],
"outputs": [
{
"name": "samples",
"type": "LATENT",
"links": [
193
],
"slot_index": 0
}
],
"properties": {
"Node name for S&R": "CogVideoImageEncode"
},
"widgets_values": [
false,
0
]
},
{
"id": 83,
"type": "CogVideoSampler",
"pos": {
"0": 1089,
"1": 316
},
"size": [
330,
570
],
"flags": {},
"order": 11,
"mode": 0,
"inputs": [
{
"name": "model",
"type": "COGVIDEOMODEL",
"link": 186
},
{
"name": "positive",
"type": "CONDITIONING",
"link": 183
},
{
"name": "negative",
"type": "CONDITIONING",
"link": 184
},
{
"name": "samples",
"type": "LATENT",
"link": null,
"shape": 7
},
{
"name": "image_cond_latents",
"type": "LATENT",
"link": 193,
"shape": 7
},
{
"name": "context_options",
"type": "COGCONTEXT",
"link": null,
"shape": 7
},
{
"name": "controlnet",
"type": "COGVIDECONTROLNET",
"link": null,
"shape": 7
},
{
"name": "tora_trajectory",
"type": "TORAFEATURES",
"link": null,
"shape": 7
},
{
"name": "fastercache",
"type": "FASTERCACHEARGS",
"link": null,
"shape": 7
},
{
"name": "num_frames",
"type": "INT",
"link": 195,
"widget": {
"name": "num_frames"
}
}
],
"outputs": [
{
"name": "samples",
"type": "LATENT",
"links": [
189
]
}
],
"properties": {
"Node name for S&R": "CogVideoSampler"
},
"widgets_values": [
49,
50,
6,
0,
"fixed",
"CogVideoXDDIM",
1
]
},
{
"id": 58,
"type": "ImageConcatMulti",
"pos": {
"0": 1545,
"1": 679
},
"size": {
"0": 210,
"1": 150
},
"flags": {},
"order": 13,
"mode": 0,
"inputs": [
{
"name": "image_1",
"type": "IMAGE",
"link": 135
},
{
"name": "image_2",
"type": "IMAGE",
"link": 190
}
],
"outputs": [
{
"name": "images",
"type": "IMAGE",
"links": [
126
],
"slot_index": 0,
"shape": 3
}
],
"properties": {},
"widgets_values": [
2,
"right",
true,
null
]
}
],
"links": [
[
54,
20,
0,
30,
0,
"CLIP"
],
[
126,
58,
0,
59,
0,
"IMAGE"
],
[
128,
30,
0,
60,
0,
"*"
],
[
129,
60,
0,
59,
2,
"STRING"
],
[
130,
37,
0,
61,
0,
"IMAGE"
],
[
135,
61,
0,
58,
0,
"IMAGE"
],
[
150,
59,
0,
44,
0,
"IMAGE"
],
[
152,
72,
0,
65,
2,
"INT"
],
[
173,
65,
0,
80,
0,
"IMAGE"
],
[
174,
80,
0,
37,
0,
"IMAGE"
],
[
183,
30,
0,
83,
1,
"CONDITIONING"
],
[
184,
31,
0,
83,
2,
"CONDITIONING"
],
[
186,
84,
0,
83,
0,
"COGVIDEOMODEL"
],
[
188,
84,
1,
85,
0,
"VAE"
],
[
189,
83,
0,
85,
1,
"LATENT"
],
[
190,
85,
0,
58,
1,
"IMAGE"
],
[
191,
84,
1,
86,
0,
"VAE"
],
[
192,
61,
0,
86,
1,
"IMAGE"
],
[
193,
86,
0,
83,
4,
"LATENT"
],
[
194,
30,
1,
31,
0,
"CLIP"
],
[
195,
72,
0,
83,
9,
"INT"
]
],
"groups": [],
"config": {},
"extra": {
"ds": {
"scale": 0.6303940863129809,
"offset": [
814.9475817376318,
180.21736528457424
]
}
},
"version": 0.4
}