Reposer
5.0
1 reviewsDescription
(This template is used for Workflow Contest)
What this workflow does
This workflow is designed to create a character in any pose with a consistent face, based on a single face input face image and an image of the pose required for the character.
How to use this workflow
1. Input Face Image (optionally change IPAdapter settings)
2. Input Pose Image (optionally change pose settings)
3. Prompt
4. Select Initial Model Option (model2 should be the default, or model1 for "creative" mode)
Tips about this workflow
Use the IPAdapter recommendations for the input face image (face only, white background) if possible, though it works well without. For the most part you just pick your two images, prompt and render!
🎥 Video demo link (optional)
This is an updated version, but for a video of the original version - https://youtu.be/SacK9tMVNUA
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (1)
- latest (7 months ago)
Node Details
Primitive Nodes (11)
Anything Everywhere (3)
Anything Everywhere3 (1)
Anything Everywhere? (1)
IPAdapterApply (1)
Note (3)
Prompts Everywhere (1)
Reroute (1)
Custom Nodes (41)
- CR Model Input Switch (1)
ComfyUI
- CheckpointLoaderSimple (1)
- ControlNetLoader (1)
- PreviewImage (4)
- ConditioningCombine (1)
- VAELoader (1)
- ControlNetApplyAdvanced (1)
- EmptyLatentImage (1)
- VAEDecode (3)
- CLIPTextEncode (3)
- FreeU_V2 (1)
- CLIPVisionLoader (1)
- UpscaleModelLoader (1)
- KSampler (2)
- SaveImage (3)
- ImageScaleToTotalPixels (1)
- LoadImage (2)
- KSamplerAdvanced (1)
- ImpactSegsAndMask (1)
- SAMLoader (1)
- UltralyticsDetectorProvider (1)
- SAMDetectorCombined (1)
- BboxDetectorSEGS (1)
- DetailerForEachDebug (1)
- DWPreprocessor (1)
- IPAdapterModelLoader (1)
- PrepImageForClipVision (1)
- NNLatentUpscale (2)
- GetImageSize (1)
Model Details
Checkpoints (1)
SD1.5/sd_15_realcartoon3d_v8.safetensors
LoRAs (0)