【AI辅助设计】完成进化!ComfyUI-Anyline正式发布

部署运行你感兴趣的模型镜像


ComfyUI-Anyline作为一款先进的AI辅助设计工具,已经完成进化,正式发布。本文将介绍ComfyUI-Anyline的核心特点和应用,以及它如何为您的设计工作带来更高的效率和便利。
前言

前段时间,相信大家都知道MistoLine 的发布了,核心的意义在于:一种比canny和其他线条控制方法更加准确、细致的线条controlnet控制。看看官方的对比图就知道了:

preprocessors.png

网上有太多关于MistoLine的介绍了,我认为应用在工作流上,还缺预处理器这里。

就在刚刚,作者发布了ComfyUI-Anyline,终于完成进化了!

应用场景

适合用于之前Canny等线条控制的一切应用场景,包括但不限于:

  • 绘画领域。线稿生成特定风格的色彩图

  • 建筑领域。线稿、草模生成渲染效果图

  • 产品设计领域。线稿、草模生成产品渲染效果图

  • 营销设计。产品图生成场景图。这是工作流的部分,可以替代Canny部分。

  • 视频/图片风格转绘。将视频或图片转绘为其他风格

Anyline介绍

Anyline: 快速、精确、详细的线条检测预处理器

Anyline 是一个 ControlNet 线条预处理器,能够从大多数图像中准确地提取对象边缘、图像细节和文本内容。用户可以输入任何类型的图像,快速获得线条清晰、细节保留充分、文字高保真的线稿,这些线稿随后用作 Stable Diffusion 中的条件生成输入。

技术细节

Anyline 使用的模型和算法基于 “Tiny and Efficient Model for the Edge Detection Generalization (TEED)” 论文(arXiv:2308.06468)中的创新努力。TEED 预设在 ComfyUI 中也源自这项工作,标志着它是一个强大的视觉算法(TEED 目前是最先进的)。更多细节,请参考论文。

与其他线条提取预处理器的比较

Anyline 使用 1280px 的处理分辨率,因此比较也是在这一分辨率下进行的。与其他常用的线条预处理器相比,Anyline 在轮廓精度、对象细节、材质纹理和字体识别(特别是在大型场景中)方面具有显著优势。它还在减少大多数场景中的噪声方面表现更好,从而在生成过程中实现更干净的图像处理和更少的不准确之处。

效果概览

Anyline 与 Mistoline ControlNet 模型结合,形成了一个完整的 SDXL 工作流程,最大限度地实现了精确控制并利用了 SDXL 模型的生成能力。Anyline 也可以在 SD1.5 工作流程中与 SD1.5 的 ControlNet 一起使用,尽管它在 SDXL 工作流程中的 Anyline+MistoLine 设置中通常表现更好。

注意:最终结果高度依赖于所使用的基础模型。请根据需要选择合适的基础模型。

安装

要将 Anyline 作为 ComfyUI 插件使用,您需要先安装 comfyui_controlnet_aux!您可以按照这里的说明进行操作:https://github.com/Fannovel16/comfyui_controlnet_aux?tab=readme-ov-file#installation

安装完 comfyui_controlnet_aux 后,请按照以下步骤操作:

  1. 打开您的 IDE 和 ComfyUI 命令行,确保 Python 环境匹配。

  2. 导航到自定义节点目录:
    cd custom_nodes

  3. 克隆存储库,或手动下载此存储库并放入 ComfyUI/custom_nodes 目录:
    git clone https://github.com/TheMistoAI/ComfyUI-Anyline.git

  4. 导航到克隆的目录:
    cd ComfyUI-Anyline

  5. 安装依赖项
    pip install -r requirements.txt

    首次使用时,模型将自动下载。如果失败,请手动从 HuggingFace Repo 下载,并把 .pth 文件放置在指定目录。

实测效果

搭建测试工作流

我们直接跟Canny做对比,使用相同的模型和提示词,最终输出对比图。我们做两个对比,一个是对比Canny,分别的出图质量,一个是使用Anyline在不同contorlnet权重和降噪下,出图效果对比。

跟Canny对比

工作流如下:

公共参数

Anyline

Canny

对比输出

测试对比结果

科幻人物

风景

工业产品

动漫IP

艺术绘画

写实人像摄影

建筑类

测试结果看,Anyline是全面超越了Canny,大家可以放心冲了!!

Anyline不同参数对比

搭建工作流

使用效率节点,快速搭建XY对比图:

对比结果

我们输入一张原图。

如下图,我们得到最佳的参数组合是:

image 2_compress.jpg

降噪0.9附近,controlnet强度在1,这套参数组合,可以很好还原人物特征,同时保留一定的自由度。

资料软件免费放送

次日同一发放请耐心等待

关于AI绘画技术储备

学好 AI绘画 不论是就业还是做副业赚钱都不错,但要学会 AI绘画 还是要有一个学习规划。最后大家分享一份全套的 AI绘画 学习资料,给那些想学习 AI绘画 的小伙伴们一点帮助!

感兴趣的小伙伴,赠送全套AIGC学习资料和安装工具,包含AI绘画、AI人工智能等前沿科技教程,模型插件,具体看下方。

需要的可以扫描下方优快云官方认证二维码免费领取【保证100%免费】

在这里插入图片描述

**一、AIGC所有方向的学习路线**

AIGC所有方向的技术点做的整理,形成各个领域的知识点汇总,它的用处就在于,你可以按照下面的知识点去找对应的学习资源,保证自己学得较为全面。

在这里插入图片描述

在这里插入图片描述

二、AIGC必备工具

工具都帮大家整理好了,安装就可直接上手!
在这里插入图片描述

三、最新AIGC学习笔记

当我学到一定基础,有自己的理解能力的时候,会去阅读一些前辈整理的书籍或者手写的笔记资料,这些笔记详细记载了他们对一些技术点的理解,这些理解是比较独到,可以学到不一样的思路。
在这里插入图片描述
在这里插入图片描述

四、AIGC视频教程合集

观看全面零基础学习视频,看视频学习是最快捷也是最有效果的方式,跟着视频中老师的思路,从基础到深入,还是很容易入门的。

在这里插入图片描述

五、实战案例

纸上得来终觉浅,要学会跟着视频一起敲,要动手实操,才能将自己的所学运用到实际当中去,这时候可以搞点实战案例来学习。
在这里插入图片描述
这份完整版的学习资料已经上传优快云,朋友们如果需要可以微信扫描下方优快云官方认证二维码免费领取【保证100%免费】

在这里插入图片描述

您可能感兴趣的与本文相关的镜像

ComfyUI

ComfyUI

AI应用
ComfyUI

ComfyUI是一款易于上手的工作流设计工具,具有以下特点:基于工作流节点设计,可视化工作流搭建,快速切换工作流,对显存占用小,速度快,支持多种插件,如ADetailer、Controlnet和AnimateDIFF等

After restarting ComfyUI, please refresh the browser. translate\translate.go:25:2: github.com/abadojack/whatlanggo@v1.0.1: Get "https://proxy.golang.org/github.com/abadojack/whatlanggo/@v/v1.0.1.zip": dial tcp 142.251.33.81:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. translate\translate.go:28:2: github.com/andybalholm/brotli@v1.1.0: Get "https://proxy.golang.org/github.com/andybalholm/brotli/@v/v1.1.0.zip": dial tcp 142.251.33.81:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. translate\translate.go:26:2: github.com/imroc/req/v3@v3.48.0: Get "https://proxy.golang.org/github.com/imroc/req/v3/@v/v3.48.0.zip": dial tcp 142.251.33.81:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. translate\translate.go:29:2: github.com/tidwall/gjson@v1.14.3: Get "https://proxy.golang.org/github.com/tidwall/gjson/@v/v1.14.3.zip": dial tcp 142.251.33.81:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. service\service.go:23:2: github.com/gin-contrib/cors@v1.6.0: Get "https://proxy.golang.org/github.com/gin-contrib/cors/@v/v1.6.0.zip": dial tcp 142.251.33.81:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. service\service.go:24:2: github.com/gin-gonic/gin@v1.9.1: Get "https://proxy.golang.org/github.com/gin-gonic/gin/@v/v1.9.1.zip": dial tcp 142.251.33.81:443: connectex: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. [ComfyUI-Manager] Failed to perform initial fetching 'custom-node-list.json': Traceback (most recent call last): File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 347, in _wait await waiter asyncio.exceptions.CancelledError The above exception was the direct cause of the following exception: Traceback (most recent call last): File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\ComfyUI\custom_nodes\comfyui-manager\glob\manager_server.py", line 1729, in get_cache json_obj = await manager_util.get_data(uri, True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\ComfyUI\custom_nodes\comfyui-manager\glob\manager_util.py", line 146, in get_data json_text = await resp.text() ^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\client_reqrep.py", line 1260, in text await self.read() File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\client_reqrep.py", line 1218, in read self._body = await self.content.read() ^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 418, in read block = await self.readany() ^^^^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 440, in readany await self._wait("readany") File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 346, in _wait with self._timer: File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\helpers.py", line 685, in __exit__ raise asyncio.TimeoutError from exc_val TimeoutError [ComfyUI-Manager] Failed to perform initial fetching 'extension-node-map.json': Traceback (most recent call last): File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 347, in _wait await waiter asyncio.exceptions.CancelledError The above exception was the direct cause of the following exception: Traceback (most recent call last): File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\ComfyUI\custom_nodes\comfyui-manager\glob\manager_server.py", line 1729, in get_cache json_obj = await manager_util.get_data(uri, True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\ComfyUI\custom_nodes\comfyui-manager\glob\manager_util.py", line 146, in get_data json_text = await resp.text() ^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\client_reqrep.py", line 1260, in text await self.read() File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\client_reqrep.py", line 1218, in read self._body = await self.content.read() ^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 418, in read block = await self.readany() ^^^^^^^^^^^^^^^^^^^^ File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 440, in readany await self._wait("readany") File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\streams.py", line 346, in _wait with self._timer: File "C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\python\Lib\site-packages\aiohttp\helpers.py", line 685, in __exit__ raise asyncio.TimeoutError from exc_val TimeoutError FETCH ComfyRegistry Data: 5/85 FETCH ComfyRegistry Data: 10/85 FETCH ComfyRegistry Data: 15/85 FETCH ComfyRegistry Data: 20/85 FETCH ComfyRegistry Data: 25/85 FETCH ComfyRegistry Data: 30/85 FETCH ComfyRegistry Data: 35/85 FETCH ComfyRegistry Data: 40/85 FETCH ComfyRegistry Data: 45/85 FETCH ComfyRegistry Data: 50/85 FETCH ComfyRegistry Data: 55/85 FETCH ComfyRegistry Data: 60/85 FETCH ComfyRegistry Data: 65/85 FETCH ComfyRegistry Data: 70/85 FETCH ComfyRegistry Data: 75/85 FETCH ComfyRegistry Data: 80/85 FETCH ComfyRegistry Data: 85/85 FETCH ComfyRegistry Data [DONE] [ComfyUI-Manager] default cache updated: https://api.comfy.org/nodes FETCH DATA from: https://raw.githubusercontent.com/ltdrdata/ComfyUI-Manager/main/custom-node-list.json[ComfyUI-Manager] Due to a network error, switching to local mode. => custom-node-list.json => FETCH DATA from: C:\Documents\Ai\ComfyUI-aki\ComfyUI-aki-v1.6\ComfyUI\custom_nodes\comfyui-manager\custom-node-list.json [DONE] [ComfyUI-Manager] All startup tasks have been completed. [Crystools ERROR] For some reason, pynvml is not working in a laptop with only battery, try to connect and turn on the monitor [Crystools ERROR] Monitor of GPU is turning off (not on UI!) [BizyAir] Fail to get notification unread counts: Cannot connect to host api.bizyair.cn:443 ssl:default [getaddrinfo failed] 我在comfyUI中安装nunchaku有些问题 这个应该怎么修复呢
05-14
<think>好的,用户遇到了在ComfyUI中安装nunchaku时的网络连接和依赖下载问题,需要解决方案。首先,我需要回忆常见的依赖安装问题可能的原因和解决办法。常见的包括网络超时、代理设置、镜像源更换、依赖冲突等。 用户提到网络超时和下载失败,可能和网络连接不稳定有关。国内用户有时访问国外的PyPI源速度较慢,所以更换镜像源是常见的解决方法。可以建议用户使用清华源或者阿里云源,通过pip的-i参数指定镜像源安装。 另外,ComfyUI可能依赖特定版本的Python包,如果版本不兼容也会导致安装失败。这时候可以尝试升级pip,或者指定依赖的版本号。用户可能需要手动下载whl文件安装,特别是如果某些包在镜像源中没有预编译版本的话。 还有可能是系统缺少某些编译依赖,比如在安装需要编译的包时,可能需要安装gcc、python3-dev等工具。对于Windows用户,可能需要安装Visual Studio Build Tools。 用户的环境问题,比如ComfyUI的版本和节点的依赖不兼容,也可能导致安装失败。建议用户检查ComfyUI的版本是否符合nunchaku的要求,并更新到最新版本。此外,防火墙或安全软件可能会拦截下载,临时禁用这些可能有助于解决问题。 最后,如果所有方法都无效,可能需要联系开发者或查看社区是否有已知的问题和解决方案。需要将这些步骤分点整理,结构清晰,方便用户逐步排查。同时,确保提到的命令正确,镜像源链接有效,避免用户操作时出错。</think>### ComfyUI安装nunchaku依赖失败解决方案 #### 1. 网络优化方案 - **切换PyPI镜像源**(适用于国内网络环境): ```bash pip install -i https://pypi.tuna.tsinghua.edu.cn/simple [包名] --trusted-host pypi.tuna.tsinghua.edu.cn ``` 推荐镜像源:清华源/阿里云源/中科大源[^1] - **手动下载依赖包**: 1. 访问 https://pypi.org 搜索对应包名 2. 下载`.whl`格式文件 3. 本地安装: ```bash pip install /path/to/package.whl ``` #### 2. 环境修复方案 ```bash # 升级pip工具 python -m pip install --upgrade pip # 清除缓存后重试 pip install --no-cache-dir -r requirements.txt ``` #### 3. 系统级配置(Windows示例) 1. 安装C++编译工具链: - 下载Visual Studio Build Tools:https://visualstudio.microsoft.com/visual-cpp-build-tools/ - 选择安装"C++桌面开发"组件 2. 设置环境变量: ```powershell setx /M HTTP_PROXY http://127.0.0.1:1080 # 如有代理需配置 ``` #### 4. ComfyUI专用解决方案 通过`custom_nodes`目录安装: ```bash cd ComfyUI/custom_nodes/nunchaku python -m pip install -r requirements.txt --user ```
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
红包 添加红包
表情包 插入表情
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值