AI与艺术完美融合！探索写实风格绘图与实拍图的魅力

前言：

作为选手，我迫切期待着参加 #AI摄影大赛#。这个比赛是展示我的创意和技术的平台，也是与同行业互动学习的绝佳机会。在数字时代，我深知人工智能对艺术创作的重要性。

将AI辅助创作融入写实风格绘图和实拍摄影，对我来说是突破传统束缚的尝试。我曾坚守传统手法，但现在，随着人工智能技术的飞速发展，我意识到这是展现多样性和广度的机会。借助人工智能的力量，我能以高效精准的方式实现创作构想，超越限制，创造令人惊叹的作品。

工具：

智思Ai助手、Midjourney、Stable Diffusion

样图：

图片详情：

咒语：bj_Alice,1girl,Clean face,Long black hair,jewelry,flower,earrings,parted_lips,makeup,rose,bird,white headwear,white_flower,building,portrait,blue flower,yellow flower,hat flower,red lips,orange flower,cinematic lighting,strong contrast,romantic realism style,high level of detail,(dreamlike scenes),<lora:bj_Alice:0.7>,

负面提示：(ng_deepnegative_v1_75t:1.2),(badhandv4:1),(worst quality:2),(low quality:2),(normal quality:2),lowres,bad anatomy,bad hands,watermark,moles,toe,bad-picture-chill-75v,realisticvision-negative-embedding,Flowers on the face,

采样：DPM++ 2M SDE Karras

模型：亚洲人物（写实）_majicmixRealistic_v4

图片详情：

咒语：bj_Alice,1girl,full body,jewelry,flower,rose,bird,building,blue flower,yellow flower,hat flower,red lips,orange flower,cinematic lighting,strong contrast,romantic realism style,high level of detail,(dreamlike scenes),<lora:bj_Alice:0.7>,

负面提示：(ng_deepnegative_v1_75t:1.2),(badhandv4:1),(worst quality:2),(low quality:2),(normal quality:2),lowres,bad anatomy,bad hands,watermark,moles,toe,bad-picture-chill-75v,realisticvision-negative-embedding,

采样：DPM++ 2M SDE Karras

模型：亚洲人物（写实）_majicmixSombre_v20

图片详情：

咒语：Nonedpi(72.009, 72.009)XML:com.adobe.xmp<?xpacket begin=&quot;&quot; id=&quot;W5M0MpCehiHzreSzNTczkc9d&quot;?> <x:xmpmeta xmlns:x=&quot;adobe:ns:meta/&quot; x:xmptk=&quot;Adobe XMP Core 9.1-c001 79.a8d4753, 2023/03/23-08:56:37 &quot;> <rdf:RDF xmlns:rdf=&quot;http://www.w3.org/1999/02/22-rdf-syntax-ns#&quot;> <rdf:Description rdf:about=&quot;&quot; xmlns:xmp=&quot;http://ns.adobe.com/xap/1.0/&quot; xmlns:dc=&quot;404 Not Foundxmlns:photoshop=&quot;http://ns.adobe.com/photoshop/1.0/&quot;xmlns:xmpMM=&quot;http://ns.adobe.com/xap/1.0/mm/&quot;xmlns:stEvt=&quot;http://ns.adobe.com/xap/1.0/sType/ResourceEvent#&quot;xmlns:stRef=&quot;http://ns.adobe.com/xap/1.0/sType/ResourceRef#&quot;xmlns:tiff=&quot;http://ns.adobe.com/tiff/1.0/&quot; xmlns:exif=&quot;http://ns.adobe.com/exif/1.0/&quot;xmp:CreatorTool=&quot;Adobe Photoshop CS6 (Windows)&quot; xmp:CreateDate=&quot;2023-07-18T15:54:51+08:00&quot; xmp:ModifyDate=&quot;2023-07-18T16:15:17+08:00&quot; xmp:MetadataDate=&quot;2023-07-18T16:15:17+08:00&quot; dc:format=&quot;image/png&quot; photoshop:ColorMode=&quot;3&quot; xmpMM:InstanceID=&quot;xmp.iid:c2921867-b8c1-1246-a834-f17e8391d451&quot; xmpMM:DocumentID=&quot;adobe:docid:photoshop:044532fb-0e74-0241-b3dc-2409b79a1dbb&quot; xmpMM:OriginalDocumentID=&quot;xmp.did:DF9B4DF94125EE11A79B82870CA3560C&quot; tiff:Orientation=&quot;1&quot; tiff:XResolution=&quot;720000/10000&quot; tiff:YResolution=&quot;720000/10000&quot; tiff:ResolutionUnit=&quot;2&quot; exif:ColorSpace=&quot;1&quot; exif:PixelXDimension=&quot;1024&quot; exif:PixelYDimension=&quot;1536&quot;> <photoshop:TextLayers> <rdf:Bag> <rdf:li photoshop:LayerName=&quot;清&quot; photoshop:LayerText=&quot;清&quot;/> <rdf:li photoshop:LayerName=&quot;九&quot; photoshop:LayerText=&quot;九&quot;/> </rdf:Bag> </photoshop:TextLayers> <xmpMM:History> <rdf:Seq> <rdf:li stEvt:action=&quot;created&quot; stEvt:instanceID=&quot;xmp.iid:DF9B4DF94125EE11A79B82870CA3560C&quot; stEvt:when=&quot;2023-07-18T15:54:51+08:00&quot; stEvt:softwareAgent=&quot;Adobe Photoshop CS6 (Windows)&quot;/> <rdf:li stEvt:action=&quot;converted&quot; stEvt:parameters=&quot;from image/png to application/vnd.adobe.photoshop&quot;/> <rdf:li stEvt:action=&quot;saved&quot; stEvt:instanceID=&quot;xmp.iid:E09B4DF94125EE11A79B82870CA3560C&quot; stEvt:when=&quot;2023-07-18T16:06:28+08:00&quot; stEvt:softwareAgent=&quot;Adobe Photoshop CS6 (Windows)&quot; stEvt:changed=&quot;/&quot;/> <rdf:li stEvt:action=&quot;saved&quot; stEvt:instanceID=&quot;xmp.iid:478800a4-6841-bb4f-8e2e-908ca776ce36&quot; stEvt:when=&quot;2023-07-18T16:15:17+08:00&quot; stEvt:softwareAgent=&quot;Adobe Photoshop 24.7 (20230628.m.2223 0770f83) (Windows)&quot; stEvt:changed=&quot;/&quot;/> <rdf:li stEvt:action=&quot;converted&quot; stEvt:parameters=&quot;from application/vnd.adobe.photoshop to image/png&quot;/> <rdf:li stEvt:action=&quot;derived&quot; stEvt:parameters=&quot;converted from application/vnd.adobe.photoshop to image/png&quot;/> <rdf:li stEvt:action=&quot;saved&quot; stEvt:instanceID=&quot;xmp.iid:c2921867-b8c1-1246-a834-f17e8391d451&quot; stEvt:when=&quot;2023-07-18T16:15:17+08:00&quot; stEvt:softwareAgent=&quot;Adobe Photoshop 24.7 (20230628.m.2223 0770f83) (Windows)&quot; stEvt:changed=&quot;/&quot;/> </rdf:Seq> </xmpMM:History> <xmpMM:DerivedFrom stRef:instanceID=&quot;xmp.iid:478800a4-6841-bb4f-8e2e-908ca776ce36&quot; stRef:documentID=&quot;xmp.did:DF9B4DF94125EE11A79B82870CA3560C&quot; stRef:originalDocumentID=&quot;xmp.did:DF9B4DF94125EE11A79B82870CA3560C&quot;/> </rdf:Description> </rdf:RDF> </x:xmpmeta> <?xpacket end=&quot;r&quot;?>

Negative prompt:

Size: 1024x1536

感想：本模型是基于Chilloutmix_NiPrunedFP和XXMix9_realistiCV26的一个写实、美女模型。

素材做了点后期以及一部分的高P图（P的时候磨皮、皮肤修的比较多），以至于模型的皮肤看着很白很细，脖子上的结构也不是特别的明显。其人物特征会更偏向于XXMix9的V26。

用的时候可以把脸部细节词多写一点，本人很喜欢跑出来的人物图的眼线，然后重绘幅度建议0.45再高没测，迭代步数和高清修复可以多拉一点。

建议搭配不同的lora使用，会有很多意想不到的效果。

图片详情：

咒语：person,angel,wings,branch,cherry_blossoms,hanami,spring_\(season\),tree,petals,pink_flower,blurry,falling_petals,plum_blossoms,1girl,depth_of_field,flower,day,outdoors,skirt,solo,grass,wisteria,standing,blurry_foreground,in_tree,tube,blue_sky,petals_on_liquid,short_hair,path,river,pond,vibrantdetails,beautifulbackground,H.R.Gigerstyle,bestquality,masterpiece,illustration,anextremelydelicateandbeautiful,extremelydetailed,unity,wallpaper,finelydetail,bestquality,officialart,extremelydetailed,unity

负面提示：umbrella,illustration,pouches,sepia, painting, cartoons, sketch, (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, bad hands, normal quality, ((monochrome)), ((grayscale:1.2)), futanari, full-package_futanari, penis_from_girl, newhalf, collapsed eyeshadow, multiple eyebrows, fleckles,analog, analogphoto, signatre, logo,2 faces

采样：Euler

模型：XQX_0705_07084fp16

束语：樱天使模型具有非常迷人的胶片色、极致的光线氛围感以及独特的熬夜眼妆。触发TAG : person , angel , 分先后顺序。推荐Euler a和Euler,开启高清修复,CFG 7.5,Clip Skip 1。加wings可触发翅膀，不需要翅膀可不加。

图片详情：

咒语：masterpiece,bestquality,EastAsiangirl,rainbowandpurpleswirlingvortexescolourfulcloudyandsmokeandcolorfulbackground,(snowing)((swirlingvortexescolourfulhair)),wearwhitefurcoat,((longHair)),((front)),emotionalface,closeup,studiolight,studio,(((makeupportrait,pinkeyeshadow))),((withmanypurplerainbowswirlingvortexescolourfulsnowandfeatheronface)),

负面提示：ng_deepnegative_v1_75t, (badhandv4:1.2), (worst quality:2), (low quality:2), (normal quality:2), lowres, bad anatomy, bad hands, ((monochrome)), ((grayscale)) watermark, moles, large breast, big breast, sunshine, bright,

采样：DPM++ 2M SDE Karras

模型：追求极限Extreme Realistic

束语：更偏重写实realistic,不会太亚洲脸，欧美风格也适用。所有图片依然是直出，没有lora，就是最普通的用法。

手办风格

图片详情：

咒语：ew, 1girl, halo, solo, heart, hair over one eye, wings, blue eyes, black background, white dress, angel, looking at viewer, dress, smile, hair ornament, chibi, blush, white hair, bangs, closed mouth, long sleeves, simple background, bare shoulders, braid, angel wings, v, white wings, heart hair ornament, off shoulder, hands up, food, collarbone

负面提示：_DeepNegative_V1_75T ,sketch, duplicate, ugly, huge eyes, text, logo, monochrome, worst face, (hand,bad and mutated hands:1.3), (worst quality:2.0), (low quality:2.0), (blurry:2.0), horror, geometry, bad_prompt, (bad hands) (missing fingers), multiple limbs, bad anatomy, (boy,interlocked fingers:1.2), Ugly Fingers, (extra digit and hands and fingers and legs and arms:1.4), crown braid, ((2girl)), (deformed fingers:1.2), (long fingers:1.2), (bad-artist-anime), bad-artist, bad hand,

采样：DPM++ 2M Karras

模型：基础

束语：权重不建议到1，直接跑的话0.85，配合其他lora的话就要拉低一些了，我这个有点过拟合，主要是背光强烈点。这个lora挺有意思的，用了10多张素材就出效果了，扒的MJ的图练的。

最后结语：

参与AI摄影艺术的竞赛，让我感受到了Ai助手的巨大潜力和无限可能。通过智思AI等工具的支持，我们能够开拓创作思路，实现更多元化、创新性的作品。Ai助手不仅能提供技术支持，还能为我们带来新的灵感和思考。与Ai的结合不仅是一次科技与艺术的碰撞，更是拓展我们创作领域的新冒险。

最后需要对应模型的你，可以私信我，我会免费分享给大家！