【亚马逊云新春特辑⑤】构生成式 AI 文生图工具之借助 ControlNet 进行 AI 绘画创作【生成拜年图】

Amazon EC2
生成式人工智能
技领云博主
0
0
>文章作者:云矩阵 ### 4. 生成拜年图 本节将为大家演示如何使用 imAgine 绘图方案生成新春贺年图,以下呈现了几张效果图,祝大家龙年大吉! Stable Diffusion (SD)是2022年发布的开源的文生图模型,通用性强,但无法满足对细节和特定内容的绘图需求,因此人们通常会对官方 SD 模型进行微调来得到可以应用于特定领域的定制模型。常见的 SD 微调技术包括 Dreambooth ,LoRA ,Textual Inversion (也称为 Embedding)和 Hypernetworks 等,几种技术的对比可参考此博客 。为了方便您使用,本次实验涉及到的模型均已预装入专用的 AMI 中。 #### 4.1 实验环境准备 实验开始前,请确保您已完成以下事项: - 已按照0.简介与前提条件申请了加速计算型实例资源 - 已按照1.订阅和部署解决方案订阅并部署了imAgine绘图解决方案。 #### 4.2 图片生成 ##### 1.参数配置 ``` 1girl, eastern dragon, open mouth, smile, long sleeves, chinese clothes, black hair, brown eyes, dress, short hair, socks, simple background, hair ribbon, red ribbon, china dress, yellow eyes, red footwear, red dress,ribbon, two side up, bangs, tabi, brown hair, :d, red background<lora:mw_hbfm_v1:0.85> Negative prompt: easynegative,dark,bad hands,bad feet,worst quality,low quality,normal quality,bad artist,bad anatomy,Japanese style Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 7, Seed: 475956187, Size: 512x728, Model hash: e3edb8a26f, Model: ghostmix_v20Bakedvae, VAE hash: 735e4c3a44, VAE: vae-ft-mse-840000-ema-pruned.safetensors, Denoising strength: 0.5, Clip skip: 2, ADetailer model: face_yolov8n.pt, ADetailer prompt: "Little girl, young, Smile", ADetailer confidence: 0.3, ADetailer dilate erode: 4, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer model 2nd: hand_yolov8n.pt, ADetailer prompt 2nd: "Little girl, young, delicate hands, delicate fingers", ADetailer confidence 2nd: 0.3, ADetailer dilate erode 2nd: 4, ADetailer mask blur 2nd: 4, ADetailer denoising strength 2nd: 0.4, ADetailer inpaint only masked 2nd: True, ADetailer inpaint padding 2nd: 32, ADetailer version: 24.1.1, Hires upscale: 2, Hires steps: 10, Hires upscaler: R-ESRGAN 4x+ Anime6B, Lora hashes: "mw_hbfm_v1: 36178776ef7d", TI hashes: "easynegative: c74b4e810b03", Version: v1.7.0 ``` 首先将上述参数复制到 SD Web UI 的正向提示词文本框中,如下图所示 ![image.png](https://dev-media.amazoncloud.cn/7798f75280d34d428e2422ea0477d666_image.png "image.png") ![image.png](https://dev-media.amazoncloud.cn/44cb57665ebd4c3ba66d1090ca7373fb_image.png "image.png") ![image.png](https://dev-media.amazoncloud.cn/6ac492d050aa4cb8b600321be027a843_image.png "image.png") ![image.png](https://dev-media.amazoncloud.cn/99dc21b081404d439525830d1583e3d8_image.png "image.png") ![image.png](https://dev-media.amazoncloud.cn/0a80232f2bdb405c9fd4170ea0cef9d8_image.png "image.png") ### 总结 通过本次动手训练营相信您已经对 imAgine 解决方案有了更深入的了解。更多的参考材料以您可以参考以下链接: - Stable Diffusion AI 方案解决方案官网:https://www.ecloudrover.com/aigc/?trk=cndc-detail - imAgine操作手册:https://rain-plus-edu.notion.site/imAgine-50d4d4d73dc34a8cb088c4b84d394d33?trk=cndc-detail - Stable Diffusion AI 方案 MarketPlace 订阅链接:https://aws.amazon.com/marketplace/pp/prodview-ohjyijddo2gka?sr=0-1&ref_=beagle&applicationId=AWSMPContessa?trk=cndc-detail 如果您需要删除本次动手训练营部署的资源,您可以直接删除您的 [EC2](https://aws.amazon.com/cn/ec2/?trk=cndc-detail) 以节省成本。 ![image.png](https://dev-media.amazoncloud.cn/c19628332b40489db209f308e9d17a27_image.png "image.png") ![image.png](https://dev-media.amazoncloud.cn/d50b8783aa3f45efbfcaadec2d48c216_image.png "image.png") [![1.png](https://dev-media.amazoncloud.cn/0adeb951b1a84dc89177a9ae00538407_1.png "1.png")](https://summit.amazoncloud.cn/2024/register.html?source=DSJAVfG2GS7gEk2Osm6kYXAa+8HnSEVdbCVjkuit7lE= )
0
目录
关闭