Expression-Snapshots
4.5
4 reviewsDescription
Flow-App instruction:
This is a relatively simple Flow-App that showcases the power of A.i. photo manipulation.
🔴 1. Upload a close up portrait image.
🔴 Now press Press CREATE to see the result in a minute ! 💚
------------------------------------------------------
This workflow takes a user photo as input (Pink node)
And generates 1 image with 4 versions of the person from the input photo in four different facial expressions.
Neutral , Smiling, Sad , Angry
Enjoy !
#NeuraLunk
-------------
This workflow was based on an idea / workflow by OpenArt user @Bmad.
Credits to him for bringing this idea to my attention !!!
-------------
What I added / changed to improve this workflow:
I have added Ip Adapter for more consistency in the face.
I added a Mask generator based on segmenting by keyword.
Allso added 2 Controlnets for Depth and Canny.
Added a refiner at the end to smooth it all out.
Enjoy !
Sincerely,
Peter Lunk a.k.a. #NeuraLunk.
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (12)
- latest (2 years ago)
- v20240124-134253
- v20240124-122631
- v20240123-195122
- v20240123-190124
- v20240123-161106
- v20240123-155437
- v20240123-153308
- v20240123-152543
- v20240123-150206
- v20240123-133335
- v20240123-125638
Node Details
Primitive Nodes (10)
IPAdapterApply (1)
Note (2)
PrimitiveNode (1)
Reroute (6)
Custom Nodes (51)
- FlatLatentsIntoSingleGrid (1)
- ToStringList (1)
- CR Seed (1)
ComfyUI
- PreviewImage (8)
- Canny (1)
- CLIPTextEncode (4)
- CLIPVisionLoader (1)
- VAEDecode (1)
- KSamplerSelect (1)
- CheckpointLoaderSimple (1)
- ImageScale (1)
- ControlNetApplyAdvanced (2)
- SaveImage (1)
- KarrasScheduler (2)
- VAEEncode (1)
- RebatchLatents (1)
- LoadImage (1)
- BboxDetectorSEGS (1)
- UltralyticsDetectorProvider (1)
- DetailerForEach (1)
- MiDaS-DepthMapPreprocessor (1)
- AB SamplerCustom (experimental) (1)
- IPAdapterModelLoader (1)
- ControlNetLoaderAdvanced (2)
- String (4)
- ConcatStringSingle (1)
- GrowMaskWithBlur (1)
- GroundingDinoSAMSegment (segment anything) (1)
- SAMModelLoader (segment anything) (1)
- GroundingDinoModelLoader (segment anything) (1)
- Text Multiline (1)
- Convert Masks to Images (1)
- Bus Node (3)
Model Details
Checkpoints (1)
realisticVisionV51_v51VAE.safetensors
LoRAs (0)