Gabriel Jiménez

This is a workflow for creating incredible vid2vid animations utilizing a alpha mask to separate your subject and background with two separate IPAdapters! Watch video walkthrough below! The focus here was really to keep the workflow concise and straight to the point. I've noticed a trend in over engineered workflows and wanted to update / share something to help you guys get back to the most important part of what do we....CREATING! This workflow includes: • x2 video loaders (Main Video + Alpha Mask) • x2 IPAdapters (x2 Image References for Subjects + 2 Image References for Background) • Native output resolution of 512x896, 768x1344 in High Res Pass •If you need to make some alpha masks, check out the Segment Anything 2 workflow I shared that also has a video tutorial on the Civitai YouTube page! This will help you get clean masks without the need to rotoscope!Find that workflow here: <a href="https://civitai.com/models/746889/sam2-easy-comfyui-alphamattes-rotoscope" target="_blank" style="text-align: start;">https://civitai.com/models/746889/sam2-easy-comfyui-alphamattes-rotoscope</a>

jboogx UltimateLCM AnimateDiff Vid2Vid workflow! 

Flux + Lora + Upscaler

杨百万

作者微信：YJ2732639128

One click to generate Ultraman funny type pictures

cychenyue

说明文档<a href="https://xiaobot.net/post/27770e61-18df-4920-9dc9-cc4c79191d5f " target="_blank">https://xiaobot.net/post/27770e61-18df-4920-9dc9-cc4c79191d5f </a>由于AI技术更新迭代，请以文档更新为准 更多内容收录在⬇️ <a href="https://xiaobot.net/p/GoToComfyUI " target="_blank">https://xiaobot.net/p/GoToComfyUI </a>网盘 <a href="https://pan.quark.cn/s/8144fe0b40b9" target="_blank">https://pan.quark.cn/s/8144fe0b40b9</a>视频<a href="https://youtu.be/8IlmS9OnGKI" target="_blank">https://youtu.be/8IlmS9OnGKI</a>

老照片修复Flux controlNet Upscale

bole

step1: With vision model and language model, &nbsp;the description of the picture are converted into clay style prompt.step2:With flux controlnet and lora, we can transform the style of the image while maintaining similarity.step3:Crop and enlarge the face, and redraw the face to maintain similarity through sdxl and instant id models. 主要分三步：1、通过视觉模型和大语言模型，实现prompt 的风格转换2、通过Flux/lora/controlnet，在转换风格同时保持相似度。3、通过对脸部采取放大重绘再贴回，这一步使用sdxl模型和instantid模型，来恢复脸部相似度。 flux lora:https://www.liblib.art/modelinfo/9e0e6f29c0a245d3bb74bad9c7e2313e?from=personal_page&versionUuid=e0087451768a4bb3844bd084ebd79962sdxl checkpoint:https://www.liblib.art/modelinfo/36404bfb74f64c168cad048312eb5f42?from=personal_page&versionUuid=3414d7c35b1245a3b3bfcf11c980cd5bsdxl lora:https://www.liblib.art/modelinfo/235e784744304cc9833e6eaad5d9b58b?from=personal_page&versionUuid=29269f9228b84c8cad139a15c3372551 手机端H5在线使用链接（flux模型叠加sdxl模型会有点慢）：<a href="https://tt.9syun.com/app/index.php?i=66&c=entry&do=mobile&m=tech_huise&r=index.index.index#/pages/zhixing/zhixing?id=3155&tg=0&v=2" target="_blank">https://tt.9syun.com/app/index.php?i=66&c=entry&do=mobile&m=tech_huise&r=index.index.index#/pages/zhixing/zhixing?id=3155&tg=0&v=2</a>

OpenArt

Workflows

All Workflows

Active Sessions

All ComfyUI Workflows

Loading...