Can you make a workflow that generates consistent characters in different poses? You have UNTIL SUNDAY to participate ! 
Info on the #Challenges Discord Channel. WIN an Amazon Gift-Card! 

Full Character Consistancy! - CLOSED

#NeuraLunk

My attempt at Full Character Consistancy...Although 100% consistency may be almost impossible to achieve with SD (yet).A work in progress ;) I generated 4 portrait faces with another basic workflow.These are then all fed into IPAdapter FaceID plus v2 SDXL.(See resource download links on the right of this page)To Keep the same Character in different poses consistant,you need to keep the seed and prompt fixed when you have found a good prompt with good results.Then you can change the character's pose with an image and the OpenPose controlnet... (I will add this soon)... #NeuraLunk

An attempt at Full-Character Consistancy. (Using SDXL model with Lightning 8-step lora)

Aderek

It's not easy to create several photos with the same character... This version works with SDXL models (recommended &nbsp;<a href="https://civitai.com/models/237678?modelVersionId=351046" target="_blank">https://civitai.com/models/237678?modelVersionId=351046</a> ) on an RTX 3060ti GPU or better. At the beginning, you upload a startup image (size about 1024x1024) and enter the path to a folder with six images, from which the system will load poses. The system will then create six images and proceed to try to apply identical outfits to the character. The better models you have access to detect items of clothing, the better the effect will be. This set only creates images in the basic form, i.e. if you want to get better results, you need to add e.g. a face detailer. Have fun with this PE. Financial support welcome ;) <a href="https://ko-fi.com/aderek514" target="_blank">Support Paweł Tomczuk on Ko-fi! ❤️. </a><a href="http://ko-fi.com/aderek514" target="_blank">ko-fi.com/aderek514</a><a href="https://ko-fi.com/aderek514" target="_blank"> - Ko-fi ❤️ Where creators get support from fans through donations, memberships, shop sales and more! The original 'Buy Me a Coffee' Page.</a> Visit my DeviantArt page: <a href="https://www.deviantart.com/aderek" target="_blank">Aderek - Hobbyist, Digital Artist | DeviantArt</a>

IP Adapter Character Consistancy, part II

aigc101

This is a instant id workflow, I put controlnet, ipadapter, visualised prompt in. Thank you matt3o, you can find all models you need in :<a href="https://github.com/cubiq/ComfyUI_InstantID/tree/main" target="_blank">https://github.com/cubiq/ComfyUI_InstantID/tree/main</a> Have fun!

instantid-ult

Lord Lethris

<h2>What this workflow does</h2> This workflow will create a number of Character Concept Images that you can then save off and use in your own workflows. This workflow also includes an example of how you can use and maintain some form of Consistency of your character. &nbsp;Costumes will never be 100% as AI will always have creative freedom, but it's as close as I can get it. <h2>How to use this workflow</h2> This workflow requires <a href="https://github.com/comfyanonymous/ComfyUI/tree/8dc19e40d129c8ee049be7be2657458509717ba5" target="_blank">ComfyUI Revision: 2144 [8dc19e40]</a> or higher as it makes use of the Grouping feature to minimise the "Mess". 🟪 ComfyUI IPAdapter plus FaceID 2 Setup 🟪 This workflow makes use of the new FaceID 2 Models from ComfyUI IPAdapter plus - which can be found here → <a href="https://github.com/cubiq/ComfyUI_IPAdapter_plus" target="_blank">https://github.com/cubiq/ComfyUI_IPAdapter_plus</a> ← and may need setting up in a handful of Nodes 🟦 Other Settings 🟦 All other custom configuration is in 🟦 Blue 🟦.Most of these are self explanatory, especially if you are ComfyUI Savvy.However I do recommend the following settings (I've highlighted with a 🟡) as this is what the workflow was optimised using: 🟡 BBOX / SEGM - Face Models: &nbsp; &nbsp; Model_Name &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;= face_yolov8n_v2.pt &nbsp; &nbsp; SEGM Face Model = face_yolov8n_seg2_60.pt 🟡 SAM &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; = sam_vit_h_4b8939pth🟡 Face Restore Model &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;= codeformer.pth or RestoreFormer.pth🟡 Load InsightFace &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;= CPU preferred, but whatever 😉🟡 IP-Adapter Encoder (Clip Vision) = IP_Adapter_Encoder_SD.safetensors🟡 IPAdapter Plus Model &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; = ip-adapter_sd15.bin 🟡 IP-Adapter Face Plus v2 Model &nbsp; &nbsp; &nbsp;Preset &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; = FACEID PLUS V2 &nbsp; &nbsp; &nbsp;lora_Strength = 0.80 -&gt; 0.90 &nbsp; &nbsp; &nbsp;Provider &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;= CPU preferred, but whatever floats Your Boat 😉 There are a few switches laying about (I've highlighted with a ⚪) that you can use to switch between random content and your own prompt, and a style switch that helps define reality from Anime.The Concept Art Face Fixer Node will take a while - this step helps improve the facial Consistency of the character.I have added Clothing Tweaks. &nbsp;The Default Settings should be good, but feel free to play around with this. 🟪 DEBUG Switch 🟪 Debugging Switch - Set This to "2", and then "Right click -&gt; Bypass" on Both Upscale Quality Fix and Concept Art Face Fixer - To speed up workflow but compromise quality. 🟫 Concept Art Pose 🟫 Here you need to load the Open Pose imageYou can download this from the Resource section on the left. <img src="https://cdn.openart.ai/wangeditor_uploads/MFnwda4rIljJ8vgEAFMu/compressed/image_0t5GW8rD_1708637287686_512.webp" alt="" data-href="" style=""/> <h2>Tips about this workflow</h2> Try to keep custom prompts to a minimum, as the workflow will add extra "stuff" to the prompts for optimal Character creation. I have found this workflow to work really well in Anime mode switched on, and with Complementary Anime based Checkpoints and with a "CLIP Skip" of -2 and "steps" of about 15-20. <img src="https://cdn.openart.ai/wangeditor_uploads/MFnwda4rIljJ8vgEAFMu/compressed/image_7OX5TXFh_1708642708250_512.webp" alt="" data-href="" style=""/> I will, at some point, create a companion workflow to work along side this one, and possibly create a lite version of this workflow, with just the Concept Character Creator portion. <h2>Version Changes</h2> v1.6<ul><li>Rebuilt IP Adapter Processes due to update</li><li>Removed Re-Actor Nodes as they are now redundant due to quality of IP Adapter Nodes</li><li>Added access to DEBUG Switch (It was hidden in previous versions, but I thought it maybe useful to you lot)</li><li>Removed Fix Seed node - it wasn't working right. &nbsp;You can use the debugger now anyway.</li></ul> v1.5<ul><li>Adjusted IP Adapter for Clothing</li><li>Added Green Chroma Key ( works better 90% of the time - May bleed into clothes - I'll remove if big issue)</li><li>Added Manual Clothing Influence Tweakers</li><li>Adjusted Woman / Man influence in random Character Generator (it was heavily focused on Girl/Woman)</li></ul> v1.4<ul><li>Added over 2000 Character Generations (Credit to <a href="https://civitai.com/user/Atryda" target="_blank">Atryda</a>)</li><li>Fixed issue where Random Prompt Wasn't being "Random".</li><li>Added a more intrusive Face Fix Node, as I though the old one was a little blurry.</li></ul> v1.3<ul><li>Fixed yet Another issue with Character Concept Build Node. &nbsp;This should be it..!</li></ul> v1.2<ul><li>Fixed Another issue with Character Concept Build Node - Basically Rebuilt it.</li></ul> v1.2<ul><li>Fixed issue with Character Concept Build Node</li></ul> v1.0<ul><li>Launch Version</li></ul>

Random Character Concept Creator With Image Consistency Examples V1.6

Dominic Richer

I made a workflow to be able to do consistent character in different scene, including action pose. Using IP Adapter, i could manage to have a good consistent to the character. It's far from being perfect, but it closed enough. Some Time you have to cheat it, and use you own pose reference image, sometime you have to play a little with the mask, cause IP doesn't understand it well with interaction.. but somehow, it still work very well :) Devlopper note : You better use Green Background for character ref. It work better.. all the token IP adaptor will read will be your character and maybe 1 or 2 token &nbsp;for the green background. Green is better then white in my experience. you have to have a simple promt to help the workflow.. ex for the male character the prompt is : 1 bald man with brown apron.. for 2 character it will be : 1 Bald man with brown apron and 1 blond girl with blue dress. CLOUD SETUP There is some model you will have to change to make it work, fallow the line, when it block, just try out some model were it bug... for SD1.5 use Model.psytorch for clipmodel, FacePlus V2 for Face, IP Plus for the model for BG and Character extraction. Future plan : I plan to add more character to the scene, I don t know what is the limit..Note : Your pose picture will guide the latent image to produce your image, so use 512x512 pose picture, or the size you want the result. V3 Update all the flow, but it does the same thing.. 1 Character, 2 Character or 3 Character (Din t test the 3 character) Think are more automaticFaster..

Consistant Character Maker (Comics Strip) V3

DESRYOO

This is a face changing attempt based on <a href="https://huggingface.co/ByteDance/SDXL-Lightning/blob/main/sdxl_lightning_4step_lora.pth" target="" style="text-align: left;">sdxl_lightning_4step_lora、InstantID and IPA.</a>I think a rating below 0.4 would preserve good Facial features, but this requires some manual experimentation.I hope you can gain new inspiration from it. <a href="https://huggingface.co/ByteDance/SDXL-Lightning/blob/main/sdxl_lightning_4step_lora.pth" target="" style="text-align: left;">sdxl_lightning_4step_lora:</a> <a href="https://huggingface.co/ByteDance/SDXL-Lightning/tree/main" target="_blank" style="text-align: start;">https://huggingface.co/ByteDance/SDXL-Lightning/tree/main</a>Instantid Node: <a href="https://github.com/cubiq/ComfyUI_InstantID" target="_blank" style="text-align: start;">https://github.com/cubiq/ComfyUI_InstantID</a>Instantid Model: <a href="https://huggingface.co/InstantX/InstantID/resolve/main/ip-adapter.bin?download=true" target="_blank">https://huggingface.co/InstantX/InstantID/resolve/main/ip-adapter.bin?download=true</a><a href="https://github.com/cubiq/ComfyUI_FaceAnalysis" target="_blank" style="text-align: start;">FaceAnalysis</a> node: <a href="https://github.com/cubiq/ComfyUI_FaceAnalysis" target="_blank" style="text-align: start;">https://github.com/cubiq/ComfyUI_FaceAnalysis</a> <img src="https://cdn.openart.ai/wangeditor_uploads/od7M2Ed2nXLWaSte4WXw/compressed/image_PGZTfszN_1708614439090_512.webp" alt="" data-href="" style="">

Test For Fast Stylized Faceswap

Abhay singh

This is a workflow for change character &nbsp;pose and youyou can adjust change in cloth happen or not by increasing or decreasing ipadapter weight

OpenArt

Flow

|

Builder

Active Sessions

Workflow Challenge- Full Character Consistancy! - CLOSED
Can you make a workflow that generates consistent characters in different poses? You have UNTIL SUNDAY to participate ! Info on the #Challenges Discord Channel. WIN an Amazon Gift-Card!
2/19/2024-2/25/2024