<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Animation |</title><link>https://hyejeong.me/tags/animation/</link><atom:link href="https://hyejeong.me/tags/animation/index.xml" rel="self" type="application/rss+xml"/><description>Animation</description><generator>HugoBlox Kit (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Mon, 04 Aug 2025 00:00:00 +0000</lastBuildDate><image><url>https://hyejeong.me/media/icon_hu_ecd623f71099616.png</url><title>Animation</title><link>https://hyejeong.me/tags/animation/</link></image><item><title>MAT-Based 3D Shape Reconstruction and Animation Using Multi-View Sketches</title><link>https://hyejeong.me/projects/sketch2anim/</link><pubDate>Mon, 04 Aug 2025 00:00:00 +0000</pubDate><guid>https://hyejeong.me/projects/sketch2anim/</guid><description>&lt;h2 id="abstract"&gt;Abstract&lt;/h2&gt;
&lt;p&gt;Animating characters from sketches is an interesting but challenging task. Existing studies have proposed shape reconstruction methods or animating methods from sketches. However, they are not suitable for character design due to the lack of user controllability. In this paper, we propose a framework for reconstructing 3D shapes and creating animations from multi-view images and skeletons. Using Medial Axis Transform (MAT) representation, the shape can be reconstructed using an inverse rendering method while preserving the structure of the initial skeleton. In addition, we propose novel distance-transform style rendering and sketch style rendering methods using characteristics of MAT. Finally, we present methods for animation transfer and animation creation. Experimental results demonstrate that our framework can easily and quickly create 3D models and animation prototypes from character sketches.&lt;/p&gt;
&lt;style&gt;
.project-gif-grid {
display: grid;
grid-template-columns: repeat(2, minmax(0, 1fr));
gap: 16px;
align-items: start;
}
.project-gif-item {
display: flex;
justify-content: center;
}
.project-gif-item img {
width: 100%;
max-width: 460px;
height: auto;
border: 1px solid #e6ebf1;
border-radius: 8px;
display: block;
}
@media (max-width: 900px) {
.project-gif-grid {
grid-template-columns: 1fr;
}
.project-gif-item img {
max-width: 100%;
}
}
&lt;/style&gt;
&lt;h2 id="reconstruction"&gt;Reconstruction&lt;/h2&gt;
&lt;div class="project-gif-grid"&gt;
&lt;div class="project-gif-item"&gt;
&lt;img src="https://hyejeong.me/uploads/projects/sketch2anim/recon_cat_input.JPG" alt="Cat input" /&gt;
&lt;/div&gt;
&lt;div class="project-gif-item"&gt;
&lt;img src="https://hyejeong.me/uploads/projects/sketch2anim/recon_cat.gif" alt="Cat reconstruction" /&gt;
&lt;/div&gt;
&lt;/div&gt;
Input sketches, 2D skeletons (left) and the reconstruction result (right).
&lt;h2 id="animation"&gt;Animation&lt;/h2&gt;
&lt;div class="project-gif-grid"&gt;
&lt;div class="project-gif-item"&gt;
&lt;img src="https://hyejeong.me/uploads/projects/sketch2anim/anim_bear.gif" alt="Bear animation" /&gt;
&lt;/div&gt;
&lt;div class="project-gif-item"&gt;
&lt;img src="https://hyejeong.me/uploads/projects/sketch2anim/anim_snowman.gif" alt="Snowman animation" /&gt;
&lt;/div&gt;
&lt;/div&gt;
Animation transfer results for reconstructed bear and snowman.
&lt;h2 id="paper-korean"&gt;Paper (Korean)&lt;/h2&gt;
&lt;p&gt;
&lt;/p&gt;
&lt;!--
&lt;iframe src="https://hyejeong.me/uploads/projects/sketch2anim/paper.pdf" width="100%" height="700" style="border: 1px solid #e5e7eb; border-radius: 8px;"&gt;&lt;/iframe&gt; --&gt;
&lt;h2 id="note"&gt;Note&lt;/h2&gt;
&lt;ul&gt;
&lt;li&gt;This project was published in a Korean academic journal (KIISE Transactions on Computing Practices (KTCP)) in 2025.&lt;/li&gt;
&lt;/ul&gt;</description></item></channel></rss>