# 使用说明

# 工作流介绍：

## 功能

1. 使用Flux + Pulid + In-Context-LoRA技术，通过一张人脸照片进行参考，实现人物一致性输出。

## 流程

1. 模型加载 + flux controlnet 高清 + 放大补充细节。
2. 裁剪成4张图，每张图进行放大补充细节。
3. 放大补充细节过程，因为都有pulid模型参与，有参考图参与，所以最终输出结果图片，能够稳定相似。

## 使用说明
1. 在单独的输入区里，可以自定义配置有：
    正向提示词、负向提示词、相似度、高清强度、放大重绘强度、放大倍数等参数，一般通过微调这些参数即可。

2. 提示词的模板包含整个图的描述，还有单独4张图片的详细描述。然后用标签[]来指定，要包含以下标签
[FOUR-PANEL] 
[TOP-LEFT]
[TOP-RIGHT]
[BOTTOM-LEFT] 
[BOTTOM-RIGHT]

在每个标签下面，分别描述每张图的提示词即可。

3. 还需要注意的地方是，在拆分图像后的放大补充细节处理过程，每一张图片的提示词，也需要一一对应进行配置。

4. 这个工作流我是在4090，24G显存环境下跑的。

## 作者介绍

### part1
小豹一，AIGC技术疯狂爱好和实践者，经常探索和分享最新的一些AI技术内容。

希望和大家成为朋友，分享技术的同时一起交流AI在落地上的更多可能。

### part2
我目前经常活跃在VX群，我自己也创建了一个群，可以通过扫码加入，无商务，是本人！

    https://xuwx.oss-cn-shenzhen.aliyuncs.com/wx_group.jpg

或者加入TG群，我也创建了一个群：

    https://t.me/FaceMimicAI

### part3
或者当你无聊的时候，可以偶尔来逛一下我的B站主页：

    https://space.bilibili.com/1442116774

我也会时不时发布一些新的玩意，说不定你看了后觉得有点用，或者还能有些别的想法～

### part4
同时我也在TG上制作了一些Bot，我感觉在TG上面，有很多细节的东西还是很有意思的。

可以在上面比较容易的实现自己的很多想法，并且实现的效果还不错。

你也可以亲自来TG玩一玩，或者简单体验一下，说不定你也会觉得：噢，还不错哦～！

我弄的玩意bot的地址在这里，不用添加联系人，可以直接玩。

    https://t.me/FaceMimic_ai_portrait_bot

虽然当前实现的主要能力很小，但是整个bot的流程实现得已经好全面了，没有任何影响使用的地方了。

你有一些别的建议，也可以反馈给我，或者单纯交流实现技术，我都乐于交流和分享。

---------------------------------------------------------------------------------------------------------

# Usage Instructions

# Workflow Introduction:

## Features

1. Utilizing Flux + Pulid + In-Context-LoRA technology, achieve consistent character output by referencing a single face photo.

## Process

1. Model loading + Flux ControlNet HD + enlargement to add details.
2. Crop into four images, and for each image, perform enlargement to add details.
3. During the process of enlarging and adding details, due to the involvement of the Pulid model and the reference image, the final output image can maintain stable similarity.

## Usage Instructions
1. In the separate input area, you can customize configurations such as:
   Positive prompts, negative prompts, similarity, HD intensity, enlargement redrawing intensity, enlargement ratio, etc. Usually, fine-tuning these parameters is sufficient.

2. The template for the prompts includes the overall description of the image, as well as detailed descriptions for each of the four individual images. Then use tags [] to specify, which should include the following labels:
[FOUR-PANEL] 
[TOP-LEFT]
[TOP-RIGHT]
[BOTTOM-LEFT] 
[BOTTOM-RIGHT]

Under each tag, describe the prompt words for each image accordingly.

3. It's also important to note that in the process of enlarging and adding details after splitting the image, the prompt words for each image need to be configured correspondingly.

4. 4090， 24G

## Author Introduction

### Part 1

ShermanVV, an enthusiastic explorer and practitioner of AIGC technology, often exploring and sharing the latest AI technology content.

I hope to become friends with everyone, sharing technology while discussing more possibilities of AI implementation together.

### Part 2
I am currently very active in VX groups, and I have also created a group myself. You can join through scanning the QR code, no business involved, it’s me!

    https://xuwx.oss-cn-shenzhen.aliyuncs.com/wx_group.jpg

Or join the TG group, which I've also set up:

    https://t.me/FaceMimicAI

### Part 3
Or when you're bored, you can occasionally visit my Bilibili homepage:

    https://space.bilibili.com/1442116774

I will also post some new things from time to time, maybe after watching you'll find them useful or even get some other ideas.

### Part 4
At the same time, I have created some bots on TG, and I feel that there are many interesting details on TG.

It's quite easy to realize many of your own ideas there, and the effects achieved are pretty good.

You can also come and play on TG yourself, or simply try it out; you might think: Oh, not bad at all!

The address of the bot I made is here, no need to add contacts, just play directly.

    https://t.me/FaceMimic_ai_portrait_bot

Although the main capabilities currently implemented are small, the entire bot workflow has been realized quite comprehensively, with no aspects affecting usage.

If you have any other suggestions, feel free to provide feedback to me, or simply exchange ideas about implementation techniques; I am happy to communicate and share.