AI此刻能教你画画了
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">丰色 发自 凹非寺</p>量子位 <span style="color: black;">报告</span> | 公众号 QbitAI<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">画画新手们，<span style="color: black;">怎样</span><span style="color: black;">才可</span>画出一副像样的手绘肖像？</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">除了假以时日的练习，或许<span style="color: black;">亦</span><span style="color: black;">能够</span>借助下面这款<span style="color: black;">工具</span>！</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">首要</span>，你只需画出人脸轮廓图，例如下面<span style="color: black;">这般</span>：</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic3.zhimg.com/80/v2-66047705c01e9c2dde9981e99d08bc56_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">而后</span>，计算机就能向你提示肖像整体结构的线条，供你参考：</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic2.zhimg.com/80/v2-34580aac82a50c38afd56b5395ebd4a1_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">接着，再给出人脸细节的提示线条：</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic1.zhimg.com/80/v2-1997531662b66e9fc420c6eb6d164b5c_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">最后</span>，你就能画出<span style="color: black;">这般</span>一幅作品了：</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic3.zhimg.com/80/v2-72c904157842e0db3483de956f57095a_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">是不是比新手徒手一张白纸要强<span style="color: black;">非常多</span>？</p>
<h2 style="color: black; text-align: left; margin-bottom: 10px;">两步引导教你画肖像</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">这款<span style="color: black;">工具</span>叫做dualFace。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">它之<span style="color: black;">因此</span>能辅助绘画新手和普通用户画出像样的肖像画，是<span style="color: black;">由于</span><span style="color: black;">能够</span><span style="color: black;">按照</span>你的初始线条给出人像全局框架和局部细节的素描线条。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">你只需照着它来，<span style="color: black;">无</span>经过绘画培训的“手残党”<span style="color: black;">亦</span><span style="color: black;">能够</span>完成！</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic4.zhimg.com/80/v2-6bd42df458ec5e3be801a3681f681b33_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">在全局引导<span style="color: black;">周期</span>，dualFace<span style="color: black;">按照</span>你绘制的大概轮廓，就能从内部数据库中搜索出若干<span style="color: black;">关联</span>人像，并在画布背景上<span style="color: black;">表示</span><span style="color: black;">意见</span>的人脸轮廓线。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">在局部引导<span style="color: black;">周期</span>，dualFace利用全局<span style="color: black;">指点</span>绘制的轮廓线，用深度生成模型合成人脸图像，<span style="color: black;">而后</span>将合成结果的细节（眼睛、鼻子、嘴等）<span style="color: black;">做为</span>辅助线条给出来。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">请<span style="color: black;">重视</span>，在全局<span style="color: black;">周期</span>，人像是数据库中真实的人像。而局部<span style="color: black;">周期</span>，是合成的虚拟人像。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">为何</span>两<span style="color: black;">周期</span><span style="color: black;">不消</span>同一种人像呢？</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic2.zhimg.com/80/v2-48acfa16371c56a78ae35a30fb10c705_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">这般</span>做是为了让计算机给出的<span style="color: black;">指点</span>更具多样性！毕竟数据库里的人都是有限的，<span style="color: black;">运用</span>生成模型<span style="color: black;">能够</span>在细节区分出<span style="color: black;">更加多</span>的肖像。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">另一</span>dualFace<span style="color: black;">拥有</span>切换功能，用户<span style="color: black;">能够</span><span style="color: black;">选取</span>最<span style="color: black;">恰当</span>的局部<span style="color: black;">指点</span>；</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">倘若</span>用户认为局部<span style="color: black;">指点</span>不符合<span style="color: black;">她们</span>的想象，<span style="color: black;">亦</span><span style="color: black;">能够</span><span style="color: black;">经过</span>点击按钮来使绘制过程返回到全局<span style="color: black;">周期</span>。</p>
<div style="color: black; text-align: left; margin-bottom: 10px;"><img src="https://pic1.zhimg.com/80/v2-2a62876b01b2d613812e61bf97fcf628_720w.webp" style="width: 50%; margin-bottom: 20px;"></div>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">下面就<span style="color: black;">来讲</span>说这两个<span style="color: black;">周期</span><span style="color: black;">详细</span><span style="color: black;">怎样</span>操作的：</p>
<h2 style="color: black; text-align: left; margin-bottom: 10px;">全局引导</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">全局引导<span style="color: black;">周期</span>分为三步：数据生成，轮廓匹配和交互式<span style="color: black;">指点</span>。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">数据生成<span style="color: black;">便是</span>把数据库中的人脸图像转换成人脸轮廓。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">她们</span><span style="color: black;">运用</span>双向分割网络（BiSeNet）来生成人脸原图的语义标签蒙层（</p>
页:
[1]