<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Isaac Sim on Yeongseung's Blog</title><link>https://yeongseung.github.io/tags/isaac-sim/</link><description>Recent content in Isaac Sim on Yeongseung's Blog</description><generator>Hugo -- 0.160.1</generator><language>en</language><lastBuildDate>Mon, 05 Jan 2026 10:58:31 +0900</lastBuildDate><atom:link href="https://yeongseung.github.io/tags/isaac-sim/index.xml" rel="self" type="application/rss+xml"/><item><title>Generating Synthetic Datasets using Isaac Sim</title><link>https://yeongseung.github.io/posts/generating_synthetic_datasets_using_isaac_sim/post/</link><pubDate>Mon, 05 Jan 2026 10:58:31 +0900</pubDate><guid>https://yeongseung.github.io/posts/generating_synthetic_datasets_using_isaac_sim/post/</guid><description>&lt;p&gt;&lt;a href="https://github.com/Yeongseung/isaacsim-synthetic-data"&gt;Codes&lt;/a&gt;&lt;/p&gt;
&lt;h1 id="intro"&gt;Intro&lt;/h1&gt;
&lt;p&gt;For vision tasks in ML, we naturally need images and corresponding labels to train a model. Normally, we consider two options. searching and finding datasets we can leverage or creating datasets by taking a photos, labeling manually. The latter is rarely feasible, especially for a student or a small team, given the immense amount of time required for labeling.&lt;/p&gt;
&lt;figure class="figure-center"&gt;
&lt;img src="https://yeongseung.github.io/posts/Generating_Synthetic_Datasets_using_Isaac_Sim/GUI_setup.png" width="500"&gt;
&lt;figcaption&gt;Figure 1. Isaac Sim GUI&lt;/figcaption&gt;
&lt;/figure&gt;
&lt;p&gt;Here&amp;rsquo;s a third option, which is to utilize &lt;strong&gt;Isaac Sim&lt;/strong&gt;(simulation program from NVIDIA) to generate synthetic datasets. Initially, I was kind of skeptical about whether a vision model trained on simulated images would actually perform well in real environments. However, &lt;a href="https://developer.nvidia.com/blog/how-to-train-autonomous-mobile-robots-to-detect-warehouse-pallet-jacks-using-synthetic-data/"&gt;This post&lt;/a&gt; changed my mind. I recommend you to read it.&lt;/p&gt;</description></item></channel></rss>