Vintage Linez
5.0
0 reviews
Description
This is a vid2vid workflow that generates vintage comic style output. You can edit it for even thiccer lines by feeding the hed+pidi image merged with the og video to the vae encoder and the tiles controlnet. This gives you thicccc vintage linez at the cost of detail. Detail is really nice in this version.
Models/Loras you should easily find on civit, the controlnet_animate controlnet is here: https://huggingface.co/crishhh/animatediff_controlnet/tree/main (I'm using the normal controlnet_checkpoint.ckpt - as far as I'm aware there is no safetensors version of this sadly :/)
Video Demo: https://reddit.com/r/StableDiffusion/comments/1elvz3p/workflow_included_take_a_trip_to_vintage_city/
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (1)
- latest (a year ago)
Node Details
Primitive Nodes (6)
PrimitiveNode (3)
Reroute (3)
Custom Nodes (43)
- ADE_ApplyAnimateDiffModelSimple (1)
- ADE_LoadAnimateDiffModel (1)
- ADE_AnimateDiffSamplingSettings (1)
- ADE_StandardStaticContextOptions (1)
- ADE_UseEvolvedSampling (1)
- CheckpointLoaderSimpleWithNoiseSelect (1)
ComfyUI
- CLIPTextEncode (2)
- VAEDecode (1)
- ControlNetApplyAdvanced (4)
- CLIPSetLastLayer (1)
- FreeU_V2 (1)
- LoraLoader (4)
- VAEEncode (1)
- SamplerCustom (1)
- ImageInvert (3)
- ImageScale (1)
- ImageBlend (2)
- ApplyMSWMSAAttentionSimple (1)
- DWPreprocessor (1)
- TilePreprocessor (1)
- PiDiNetPreprocessor (1)
- HEDPreprocessor (1)
- IPAdapterUnifiedLoader (1)
- IPAdapterTiledBatch (1)
- ControlNetLoaderAdvanced (4)
- SamplerLCMCycle (1)
- LCMScheduler (1)
- VHS_VideoCombine (2)
- VHS_LoadVideo (1)
Model Details
Checkpoints (1)
comiccraft_v10.safetensors
LoRAs (4)
AnimateLCM_sd15_t2v_lora.safetensors
Valerian&Laureline-comic-style.safetensors
lcm_sd15.safetensors
vintcomstyle-v1-10.safetensors