[{"data":1,"prerenderedAt":226},["ShallowReactive",2],{"content-query-LRsRKtaO8N":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"state":13,"body":14,"_type":220,"_id":221,"_source":222,"_file":223,"_stem":224,"_extension":225},"/activities/zh/1780","zh",false,"","MSG广州·开发者分享 | 如何基于MindSpore实现二次元风格、染发、变老等功能？","8月31日19：30 的昇思MindSpore开发者分享，以StarGAN模型为主进行讲解，手把手教你如何基于昇思MindSpore实现漫画脸。","2022-08-31","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/09/06/c5485ba947bb43bfafb78c51ed5ed715.png","activities",1,{"type":15,"children":16,"toc":217},"root",[17,25,34,43,62,67,72,80,85,90,98,112,119,126,133,140,148,158,176,186,202],{"type":18,"tag":19,"props":20,"children":22},"element","h1",{"id":21},"msg广州开发者分享-如何基于mindspore实现二次元风格染发变老等功能",[23],{"type":24,"value":8},"text",{"type":18,"tag":26,"props":27,"children":28},"p",{},[29],{"type":18,"tag":30,"props":31,"children":33},"img",{"alt":7,"src":32},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/09/06/56dda588da754078913b986f1cca511e.gif",[],{"type":18,"tag":26,"props":35,"children":36},{},[37],{"type":18,"tag":38,"props":39,"children":40},"strong",{},[41],{"type":24,"value":42},"项目背景",{"type":18,"tag":26,"props":44,"children":45},{},[46,48,60],{"type":24,"value":47},"GAN，又叫生成对抗网络，是一种非监督的学习。该网络中有一个",{"type":18,"tag":38,"props":49,"children":50},{},[51,53,58],{"type":24,"value":52},"生成器（Generators）",{"type":18,"tag":38,"props":54,"children":55},{},[56],{"type":24,"value":57},"和",{"type":24,"value":59},"判别器（Discriminators）",{"type":24,"value":61},"，而训练过程就是这两个网络不断博弈对抗的过程。生成器不断生成假图企图通过判别器的识别，而判别器则将图片划分为真实图像和生成图像。",{"type":18,"tag":26,"props":63,"children":64},{},[65],{"type":24,"value":66},"◆ 在该模型中，生成器的输入是一串噪音，输出是一张生成的假图，而生成器致力于让判别器无法识别出这张假图是生成的图还是真实图像。在训练过程中，不断用判别器的分数做反馈使生成器效果越来越好。",{"type":18,"tag":26,"props":68,"children":69},{},[70],{"type":24,"value":71},"◆ 判别器的输入是一张图片，输出则是图片的分数，分数越高说明此时生成的图像越接近真实图像。判别器致力于识别图片是真图还是假图 ，在训练过程中不断投喂假图，输出一个分数再与真实图像的标签进行比较。实际上也是一个二分类的过程。",{"type":18,"tag":26,"props":73,"children":74},{},[75],{"type":18,"tag":38,"props":76,"children":77},{},[78],{"type":24,"value":79},"直播要点抢先看",{"type":18,"tag":26,"props":81,"children":82},{},[83],{"type":24,"value":84},"基于昇思MindSpore实现基于GAN网络模型的漫画脸，输入一张图片，输出一张漫画脸，可以用作手机变装，二次风格等应用场景。",{"type":18,"tag":26,"props":86,"children":87},{},[88],{"type":24,"value":89},"解决方案：基于StarGAN、AnimeGANv2，实现染发、变老、变儿童、二次元风格等功能。",{"type":18,"tag":26,"props":91,"children":92},{},[93],{"type":18,"tag":38,"props":94,"children":95},{},[96],{"type":24,"value":97},"MSG广州·开发者分享直播",{"type":18,"tag":26,"props":99,"children":100},{},[101,110],{"type":18,"tag":38,"props":102,"children":103},{},[104],{"type":18,"tag":105,"props":106,"children":107},"em",{},[108],{"type":24,"value":109},"8月31日（今晚）19：30",{"type":24,"value":111}," 的昇思MindSpore开发者分享，以StarGAN模型为主进行讲解，手把手教你如何基于昇思MindSpore实现漫画脸。",{"type":18,"tag":26,"props":113,"children":114},{},[115],{"type":18,"tag":30,"props":116,"children":118},{"alt":7,"src":117},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/09/06/0a3c54b6cd9c476eab193c5b6254ee16.jpg",[],{"type":18,"tag":26,"props":120,"children":121},{},[122],{"type":18,"tag":30,"props":123,"children":125},{"alt":7,"src":124},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/09/06/27cc6f0d8ba646f99123a8c15186ac7c.png",[],{"type":18,"tag":26,"props":127,"children":128},{},[129],{"type":18,"tag":30,"props":130,"children":132},{"alt":7,"src":131},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/09/06/2df151f3011e4c539c4753d81abdd72f.png",[],{"type":18,"tag":26,"props":134,"children":135},{},[136],{"type":18,"tag":30,"props":137,"children":139},{"alt":7,"src":138},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/09/06/e143f62ae807400fbcbb9933184d3e8a.jpg",[],{"type":18,"tag":26,"props":141,"children":142},{},[143],{"type":18,"tag":38,"props":144,"children":145},{},[146],{"type":24,"value":147},"MindSpore官方资料",{"type":18,"tag":26,"props":149,"children":150},{},[151,156],{"type":18,"tag":38,"props":152,"children":153},{},[154],{"type":24,"value":155},"官方QQ群",{"type":24,"value":157}," : 486831414",{"type":18,"tag":26,"props":159,"children":160},{},[161,166,168],{"type":18,"tag":38,"props":162,"children":163},{},[164],{"type":24,"value":165},"官网",{"type":24,"value":167},"：",{"type":18,"tag":169,"props":170,"children":174},"a",{"href":171,"rel":172},"https://www.mindspore.cn/",[173],"nofollow",[175],{"type":24,"value":171},{"type":18,"tag":26,"props":177,"children":178},{},[179,184],{"type":18,"tag":38,"props":180,"children":181},{},[182],{"type":24,"value":183},"Gitee",{"type":24,"value":185}," : https : //gitee.com/mindspore/mindspore",{"type":18,"tag":26,"props":187,"children":188},{},[189,194,196],{"type":18,"tag":38,"props":190,"children":191},{},[192],{"type":24,"value":193},"GitHub",{"type":24,"value":195}," : ",{"type":18,"tag":169,"props":197,"children":200},{"href":198,"rel":199},"https://github.com/mindspore-ai/mindspore",[173],[201],{"type":24,"value":198},{"type":18,"tag":26,"props":203,"children":204},{},[205,210,211],{"type":18,"tag":38,"props":206,"children":207},{},[208],{"type":24,"value":209},"论坛",{"type":24,"value":167},{"type":18,"tag":169,"props":212,"children":215},{"href":213,"rel":214},"https://bbs.huaweicloud.com/forum/forum-1076-1.html",[173],[216],{"type":24,"value":213},{"title":7,"searchDepth":218,"depth":218,"links":219},4,[],"markdown","content:activities:zh:1780.md","content","activities/zh/1780.md","activities/zh/1780","md",1776506008976]