[AIDv2.91] 二次元插画设计

CHECKPOINT
Original


Updated:

6

0 前言

我厌倦了传统 AI 绘画一成不变的脸、姿势、风格,所以想要脱离混合模型。最初,我使用提示词,可始终无法达成某种微妙的线条、色彩、光影、质感、构图或故事性,甚至无法复刻模型偶然产生的惊艳风格。这种昙花一现仅与一般风格有细微差别,却从美学上引人入胜。因此,我想制作一种能完美学习艺术风格并稳定输出的模型。我从 2022 年 11 月开始收集素材训练风格化模型,特殊打标以区分那些仅有细微差异的素材,终于于 2023 年年初在模型风格上自成一派,即 AIDv1.0 模型。

为什么不练 Lora 而要微调?我始终认为微调的效果要优于 Lora。它不依赖于底模,所有的训练图像在训练中共同向着误差最低点前进,而不仅是最优化一块附加权重。但我也在探寻能够将特定风格完美融入大模型的方法,以减轻训练负担。

此后半年里,我自费两万多元,亲自裁图、打标、魔改脚本。训练步数从几千,几万再到几百万,训练设备从 RTX3060,RTX3090 再到 A100。从制作素材再到训练,AID 也逐渐成为了架构完整的工程项目。

在这之中,我发现只有当模型轻微“过拟合”到原图像的噪点时,才能对风格有最佳的学习。我尝试过拟合所有风格,并使用负面 emb 学习过拟合噪点以平衡不同风格间的学习进度,由此制作了 bad, badhand 和 aid 系列。这种正则化方法为我带来了很好的结果。一个训练恰到好处的负面 emb 不仅不会破坏底模的风格,还能助长风格的特征。

随着模型迭代,我认为我逐渐达到了 SD1.5 的上限。即便是微调,那些精美插画风格独特的线条、色彩、光影、构图、故事性各具特色而难以简单的 SD1.5 模型很好地学习模仿。从欠拟合到过拟合,我始终无法得到完美的风格化特征,更何况模型同时需要最优化百种以上的艺术风格。

为此,我非常期待更加复杂的 SDXL 模型能为我带来新的突破口。

模型训练期间,我并没有将精力耗费在撰写大量提示词和混合不同风格上。有人搭配一些 Lora 和非常复杂的提示词得到了相当惊艳的结果,我非常感谢他们的创新和喜爱。

最后,感谢 @BananaCat 对本文的汉化,我很乐意与全世界的 SD 爱好者分享和交流成果。AID 模型均出于专业兴趣。如果您对更多素材处理和模型训练的工程细节感兴趣,或愿意与我分享您的训练方案,欢迎在评论区留言,我会第一时间回复。

I 介绍

AnimeIllustDiffusion (AID) 是一款预训练、非商用且多风格的动漫插画模型。它 不会生成“AI脸”。它内置大量风格,您能够使用一些 特殊的触发词(见 附录 A)来生成特定风格的图像。由于内置大量内容,AID 需要强烈的负面提示才能正常工作。一般的负面提示词(例如 low quality, bad anatomy 等)效果有限,因此,若您生成的图像中出现噪点,请搭配我提供的负面文本嵌入 [1] 使用,以消除噪声。对于版本特制负面文半嵌入,请 参阅版本信息。VAE 首选 sd-vae-ft-mse-original [5]。在 Clip Skip = 1 上使用。

AID 模型拥有 超过 200 种 稳定的 动漫插画风格 100 名动漫角色。生成风格需要的特殊提示词见附录 A。生成角色则直接使用角色名。AID 模型像一个调色板,您可以通过任意组合提示词创造出新的风格。

每个版本的 AID 各有所长,并非越新的版本越好

  • 适合第一次使用:v2.8, v2.91 - Weak, v2.10beta1

  • 有极佳创造力:v2.6, v2.7, v2.91 - Weak, v2.91 - Strong

  • 较为稳定:v2.5, v2.6, v2.8, v2.91 - Weak

  • 风格多样:v2.91 - Weak, v2.91 - Strong, v2.10beta1

本页面封面图为各 AID 版本封面图总和。本页面仅上传 AIDv2.91 Weak 版本。如果您对其他版本感兴趣,请移步:

https://civitai.com/models/16828?modelVersionId=91090

II 优点

特化二次元人物插画设计。擅长平涂、厚涂和半厚涂(伪厚涂)。艺术感的线条和色彩。构图灵活大胆,同时擅长摆拍和动态姿势。细节整齐柔和,不具有混合模型的 2.5D 质感,风格自成一派,不像 AI,更接近于手绘。

认识更多热门动漫角色,或许更利于搭配角色 Lora。

III 不足

不擅长绘制人物以外的场景。不擅长油画和水彩画风。需要搭配特制负面 emb 以消除噪点。触发词之间强度不够平衡。对自然语言的理解能力较弱,与大部分风格化 Lora 和小部分角色 Lora 不适配。

IV 声明

本模型用于测试多风格模型训练,非盈利或商用,皆兴趣使然。若有侵权,立即删除。

使用者仅被授权使用此模型生成图片,不允许未经同意的转载。

严禁将本模型用于一切商业用途。

请勿使用本模型生成带有血腥、暴力、色情的违规图片及任何侵权内容!因此,附录 A 部分仅能够提供部分经过训练的关键词。

附录 A

截止至 AIDV2.5:by 35s00, by agm, by ajimita, by akizero, by ask, by chicken utk, by demizu posuka, by dino, by fadingz, by fuzichico, by hamukukka, by hitomio16, by ichigo ame, by key999, by kooork55, by matcha, by mika pikazo, by modare, by myung yi, by naji yanagida, by nezukonezu32, by nico tine, by nikuzume, by ninev, by oda non, by palow, by qooo003, by rolua, by samip, by serie niai, by shirentutu, by sho, by silver, by sonomura00, by void, by wlop, by xilmo, by yoneyama mai, by yosk6000, by zumizumi

AIDV2.6 新增:by caaaarrot, by hinaki, by homutan, by kazari tayu, by kitada mo, by roitz, by teffish, by ukiatsuya, by yejji, by ziyun

AIDV2.7 新增:by poharo, by jnthed, by 7thknights, by some1else45, by yohan, by yomu, by tsvbvra

AIDV2.9 新增: by kkuni, by starshadowmagic, by star furu, by rella, by tukumi bis, by yumenouchi, by chon, by eku uekura, by tira27, by kuroume, by hachisan, by nounoknown, by kurige horse, by konya karasue, by noyu, by ame929, by muryou tada, by yun216, by nekojira, by nanmo, by wait ar, by akasaai, by momoco, by sushi0831, by taiki, by siki, by kinta, by hata, by anteiru, by lemoneco, by umaiyo puyoman, by freng, by rin7914, by shimanun, by hidulme, by whoisshe, by 5eyo, by cutesexyrobutts, by shiren, by omutatsu, by gesoking, by 3meiji, brushstrokes

AIDV2.9 更新: (i) by demizu posuka; (ii) 修正 by fuzichico -> by fuzichoco; (iii) 提高训练图像的分辨率至768^2; (iv) 在 skip clip = 1 上训练。

AIDV2.91 新增: impasto, pseudo-impasto, semi-realistic, concept art, flat color, celluloid

直到 AIDV2.10beta1: by 35s00, by 3meiji, by 5eyo, by 7nu, by 7thknights, by adenim, by agm, by ajimita, by akizero, by ame929, by anmi, by anteiru, by arutera, by ask, by atelier irrlicht, by bunbun, by caaaaarrot, by camu, by canking, by ccroquette, by chi4, by chicken utk, by chon, by cola, by cutesexyrobutts, by darumakarei, by dino, by dora, by dsmile9, by ei maestrl, by ekita kuro, by ekita xuan, by eku uekura, by fadingz, by fajyobore, by foomidori, by freng, by fuzichoco, by gesoking, by gomzi, by hachisan, by hakuhiru oeoe, by hamukukka, by haru, by hata, by hidulme, by hikinito0902, by hinaki, by hitoimim, by hitomio16, by hizumi, by homutan, by hotatenshi, by houk1se1, by hyatsu, by icecenya, by ichigo ame, by inoriac, by iromishiro, by iwzry, by jnthed, by joezunzun, by junsui0906, by karohroka, by kaya7hara, by kazari tayu, by killow, by kin, by kinta, by kishiyo, by kitada mo, by kkuni, by konya karasue, by kooork55, by kot rou020, by krenz, by kurige horse, by kuroume, by lalalalack, by lemoneco, by lm7, by lovelymelm, by lpmya, by mar takagi, by matcha, by matsukenmanga, by melowh, by menou, by midori xu, by mika pikazo, by misumigumi, by miv4t, by mochizukikei, by mogumo, by momoco, by momoku, by morikuraen, by mqkyrie, by muina, by munashichi, by muryou tada, by myaru, by myc0t0xin, by myung yi, by nack, by naji yanagida, by nanmo, by nardack, by narue, by nekojira, by netural, by nezukonezu32, by nico tine, by nikuzume, by nine, by nineo, by ninev, by niwa uxx, by nixeu, by noco, by noodle4cool, by nounoknown, by noyu, by oda non, by omutatsu, by onineko, by palow, by panp, by pikuson, by poharo, by poire, by potg, by pro-p, by qooo003, by rai hito, by reiko, by rella, by rhtkd, by rin7914, by roitz, by ryuseilan, by saberiii, by sais, by sakiika, by samip, by sanosomeha, by say hana, by scottie0073, by senryoko, by serie niai, by seuhyo99, by shal-e, by shimanun, by shirabii, by shiraishi kanoya, by shiren, by shirentutu, by sho, by sia, by siki, by silver, by solipsist, by some1else45, by sonomura00, by sooon, by star furu, by starshadowmagic, by starzin07, by sui 0z0, by sul, by sushi0831, by suzukasuraimu, by taiki, by takumi bis, by teffish, by tidsean, by tira27, by tsukiho tsukioka, by tsvbvra, by ttosom, by tukumi bis, by uiiv, by ukiatsuya, by umaiyo puyoman, by void, by wait ar, by walzrj, by wanke, by whoisshe, by wlop, by xilmo, by yejji, by yogisya, by yohan, by yomu, by yoneyama mai, by yosk6000, by yumenouchi, by yun216, by yunikon147, by yunsang, by ziyun, by zumoti4

Version Detail

SD 1.5
3000
AIDV2.10 是一次重大更新。它使用 两倍 于 aidv2.91 数据量的数据集大小和更高的图像品质训练得到。训练图像的分辨率由原来的768像素提高到了 1024 像素,这意味着您能够直接生成1024像素(比例可变,如768x1532)的图像而不会遭受图像畸变。同时,它有着更为平衡的风格权重。另外,aidv2.10 支持 200 余种不同插画风格(aidv2.91 为 100 种左右,较原先增加约 100 种)。 同样地,AIDV2.10 有着其专属负面文本嵌入 aid210,即封面图中的 bad17。没有它,您很可能会生成出很丑陋的图像。您可以将它放在负面提示词的第一个位置以达到预期的效果。 模型封面图均由纯文生图,即不使用任何Lora和ControlNet生成,部分经过高分辨率修复或二次图生图(以完全相同的参数)放大。

Project Permissions

    Use Permissions

  • Use in TENSOR Online

  • As a online training base model on TENSOR

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated contents

  • Use on generation services

  • Sell this model or merges

Comments

Related Posts

No posts yet
Describe the image you want to generate, then press Enter to send.